| date | title |
|---|---|
2012-02-17 20:35:55 |
Cytoscape\_3/UserManual |
The NNF format is a very simple format that unlike SIF allows the optional assignment of single nested network per node. No other node columns can be specified. There are only 2 possible line formats:
- A node "node" contained in a "network:"
network node
- 2 nodes linked together contained in a network:
network node1 interaction node2
If a network name (first entry on a line) appeared previously as a node name (in columns 2 or 4), the network will be nested in the node with the same name. Also, if a name that has been previously defined as a network (by being listed in the first column), later appears as a node name (in columns 2 or 4), the previously defined network will be nested in the node with the same name. In summary: any time a name is used as both, a network name , and a node name, this implies that the network will be nested in the node of the same name. Additionally comments may be included on all lines. Comments start with a hash mark '#' and continue to the end of a line. Trailing comments (after data lines) and entirely blank lines anywhere are also permissible. Please note that if you load multiple NNF files in Cytoscape they will be treated like a single, long concatenated NNF file! If you need to embed spaces, tabs or backslashes in a name, you must escape it by preceding it with a backslash, so that, e.g. an embedded backslash becomes two backslashes, an embedded space a backslash followed by a space etc.
Example_1 C
Example_1 network1
network1 A pp B
network1 B pp A
Example_1 C pp B
Example_2 M1
Example_2 M2
M1 A
M2 B pp C
Example_2 A pp B
Example_2 M1 im M2
Example_3 M1 im M2
Example_3 M3 im M1
Example_3 M2 im M3
Example_3 C pp M3
Example_3 M2 pp C
M1 A
M2 A pp B
M3 B pp C
Example_4 M1
root M3
M1 A pp B
M1 B pp A
Example_4 C pp B
M3 M2
M2 D
M3 E pp F
M3 D pp F
M3 D pp E
Example_4 D pp C
Example_4 A pp M2
Example_4 B pp M3
Example_4 M2 pp B
Example_5 M4
M4 D
M4 M3
M3 M2 pp C
M2 M1 pp B
M1 A
M4 C pp D
In contrast to SIF, GML is a rich graph format language supported by many other network visualization packages. The GML file format specification is available at:
It is generally not necessary to modify the content of a GML file
directly. Once a network is built in SIF format and then laid out, the
layout is preserved by saving to and loading from GML. Properties
specified in a GML file will result in a new style named
Filename.style when that GML file is loaded.
XGMML is the XML evolution of GML and is based on the GML definition. In addition to network data, XGMML contains node/edge/network column data. The XGMML file format specification is available at:
XGMML is now preferred to GML because it offers the flexibility associated with all XML document types. If you're unsure about which to use, choose XGMML.
There is a java system property "cytoscape.xgmml.repair.bare.ampersands" that can be set to "true" if you have experience trouble reading older files.
This should only be used when an XGMML file or session cannot be read due improperly encoded ampersands, as it slows down the reading process, but this is still preferable to attempting to fix such files using manual editing.
The Systems Biology Markup Language (SBML) is an XML format to describe biochemical networks. SBML file format specification is available at:
BioPAX is an OWL (Web Ontology Language) document designed to exchange biological pathways data. The complete set of documents for this format is available at:
The PSI-MI format is a data exchange format for protein-protein interactions. It is an XML format used to describe PPI and associated data. PSI-MI XML format specification is available at:
GraphML is a comprehensive and easy-to-use file format for graphs. It is based on XML. The complete set of documents for this format is available at:
Cytoscape has native support for Microsoft Excel files (.xls, .xlsx) and delimited text files. The tables in these files can have network data and edge columns. Users can specify columns containg source nodes, target nodes, interaction types, and edge columns during file import. Some of the other network analysis tools, such as igraph (), has feature to export graph as simple text files. Cytoscape can read these text files and build networks from them. For more detail, please read the Import Free-Format Tables section of the Creating Networks section.
From Cytoscape 3.1.0 on, Cytoscape supports Cytoscape.js JSON files. You can use this feature to export your network visualizations to web browsers. Cytoscape.js has two ways to represent network data, and currently both reader and writer support only the array style graph notation. For example, this network in Cytoscape:
will be exported to this JSON:
{
"elements" : {
"nodes" : [ {
"data" : {
"id" : "723",
"selected" : false,
"annotation_Taxon" : "Saccharomyces cerevisiae",
"alias" : [ "RPL31A", "RPL34", "S000002233", "ribosomal protein L31A (L34A) (YL28)" ],
"shared_name" : "YDL075W",
"SUID" : 723,
"degree_layout" : 1,
"name" : "YDL075W"
},
"position" : {
"x" : 693.0518315633137,
"y" : -49.47506554921466
},
"selected" : false
}, {
"data" : {
"id" : "726",
"selected" : false,
"annotation_Taxon" : "Saccharomyces cerevisiae",
"alias" : [ "RP23", "RPL16B", "S000005013", "ribosomal protein L16B (L21B) (rp23) (YL15)" ],
"shared_name" : "YNL069C",
"SUID" : 726,
"degree_layout" : 1,
"name" : "YNL069C"
},
"position" : {
"x" : 627.3147710164387,
"y" : -205.99251969655353
},
"selected" : false
}, {
"data" : {
"id" : "658",
"selected" : false,
"annotation_Taxon" : "Saccharomyces cerevisiae",
"alias" : [ "RPL11B", "S000003317", "ribosomal protein L11B (L16B) (rp39B) (YL22)" ],
"shared_name" : "YGR085C",
"SUID" : 658,
"degree_layout" : 2,
"name" : "YGR085C"
},
"position" : {
"x" : 804.3092778523762,
"y" : -245.6235926946004
},
"selected" : false
}, {
"data" : {
"id" : "660",
"selected" : false,
"annotation_Taxon" : "Saccharomyces cerevisiae",
"alias" : [ "KAP108", "S000002803", "SXM1" ],
"shared_name" : "YDR395W",
"SUID" : 660,
"degree_layout" : 8,
"name" : "YDR395W"
},
"position" : {
"x" : 730.8733342488606,
"y" : -157.50702317555744
},
"selected" : false
}, {
"data" : {
"id" : "579",
"selected" : false,
"annotation_Taxon" : "Saccharomyces cerevisiae",
"alias" : [ "RPL11A", "S000006306", "ribosomal protein L11A (L16A) (rp39A) (YL22)" ],
"shared_name" : "YPR102C",
"SUID" : 579,
"degree_layout" : 2,
"name" : "YPR102C"
},
"position" : {
"x" : 841.1395696004231,
"y" : -130.77909119923908
},
"selected" : false
}, {
"data" : {
"id" : "578",
"selected" : false,
"annotation_Taxon" : "Saccharomyces cerevisiae",
"alias" : [ "GRC5", "QSR1", "RPL10", "S000004065", "ribosomal protein L10" ],
"shared_name" : "YLR075W",
"SUID" : 578,
"degree_layout" : 2,
"name" : "YLR075W"
},
"position" : {
"x" : 910.3755162556965,
"y" : -217.0562556584676
},
"selected" : false
} ],
"edges" : [ {
"data" : {
"id" : "659",
"source" : "658",
"target" : "578",
"selected" : false,
"interaction" : "pp",
"shared_interaction" : "pp",
"shared_name" : "YGR085C (pp) YLR075W",
"SUID" : 659,
"name" : "YGR085C (pp) YLR075W"
},
"selected" : false
}, {
"data" : {
"id" : "661",
"source" : "658",
"target" : "660",
"selected" : false,
"interaction" : "pp",
"shared_interaction" : "pp",
"shared_name" : "YGR085C (pp) YDR395W",
"SUID" : 661,
"name" : "YGR085C (pp) YDR395W"
},
"selected" : false
}, {
"data" : {
"id" : "724",
"source" : "660",
"target" : "723",
"selected" : false,
"interaction" : "pp",
"shared_interaction" : "pp",
"shared_name" : "YDR395W (pp) YDL075W",
"SUID" : 724,
"name" : "YDR395W (pp) YDL075W"
},
"selected" : false
}, {
"data" : {
"id" : "733",
"source" : "660",
"target" : "579",
"selected" : false,
"interaction" : "pp",
"shared_interaction" : "pp",
"shared_name" : "YDR395W (pp) YPR102C",
"SUID" : 733,
"name" : "YDR395W (pp) YPR102C"
},
"selected" : false
}, {
"data" : {
"id" : "727",
"source" : "660",
"target" : "726",
"selected" : false,
"interaction" : "pp",
"shared_interaction" : "pp",
"shared_name" : "YDR395W (pp) YNL069C",
"SUID" : 727,
"name" : "YDR395W (pp) YNL069C"
},
"selected" : false
}, {
"data" : {
"id" : "580",
"source" : "578",
"target" : "579",
"selected" : false,
"interaction" : "pp",
"shared_interaction" : "pp",
"shared_name" : "YLR075W (pp) YPR102C",
"SUID" : 580,
"name" : "YLR075W (pp) YPR102C"
},
"selected" : false
} ]
}
}
And this is a sample visualization in Cytoscape.js:
Export network and table to Cytoscape.js feature in Cytoscape creates a JSON file WITHOUT style. This means that you need to export the style in a separate JSON file if you apply style to your network. Please read Style section for more details.
Interaction networks are useful as stand-alone models. However, they are most powerful for answering scientific questions when integrated with additional information. Cytoscape allows the user to add arbitrary node, edge and network information to Cytoscape as node/edge/network data columns. This could include, for example, annotation data on a gene or confidence values in a protein-protein interaction. These column data can then be visualized in a user-defined way by setting up a mapping from columns to network properties (colors, shapes, and so on). The section on Styles discusses this in greater detail.
Cytoscape offers support for importing data from delimited text and MS Excel data tables.
Sample Data Table 1
Object Key Alias SGD ID AAC3 YBR085W|ANC3 S000000289 AAT2 YLR027C|ASP5 S000004017 BIK1 YCL029C|ARM5|PAC14 S000000534
The data table file should contain a primary key column and at least one data column. The maximum number of data columns is unlimited. The Alias column is an optional feature, as is using the first row of data as column names. Alternatively, you can specify each column name from the File → Import → Table → File... user interface.
-
Select File → Import → Table → File....
-
Select a data file. The file can be either a text or Excel (.xls/.xlsx) file.
-
In the Target Table Data section, choose where to import the data to. You can choose an existing network collection, a specific network only, or you can choose to import the data to an Unassigned Table (described below).
-
Depending on what you choose in the Where to import Table Data drop-down, you will need to select a network collection or specific network. You will also need to select Importing Type, that is whether the data is node, edge or network table columns.
-
If the table is not properly delimited in the preview panel, change the delimiter in the Show Text File Import Options panel. The default delimiter is tab. This step is not necessary for Excel Workbooks.
-
By default, the first column is designated as the primary key. Change the key column if necessary.
-
Click OK to import.
As of Cytoscape 3.1. it is possible to import data tables without assigning them to existing networks, meaning the data doesn't have to correspond to any nodes/edges currently loaded. If a data table is imported as unassigned and a network is later imported that maps to the data in terms of nodes or edges, the data will link automatically. This is useful when loading a large dataset (for example expression data), defining a Style for visualizing the data on networks and later loading individual networks to view the data, for example from an online database. This feature allows the data to be automatically linked to any network that is applicable, without having to load the data for each network.
In addition to tabular data, the simple attribute file format used in previous versions of Cytoscape is still supported. Node and edge data files are simply formatted: a node data file begins with the name of the column on the first line (note that it cannot contain spaces). Each following line contains the name of the node, followed by an equals sign and the data value. Numbers and text strings are the most common data types. All values for a given column must have the same type. For example:
FunctionalCategory
YAL001C = metabolism
YAR002W = apoptosis
YBL007C = ribosome
An edge data file has much the same structure, except that the name of the edge is the source node name, followed by the interaction type in parentheses, followed by the target node name. Directionality counts, so switching the source and target will refer to a different (or perhaps non-existent) edge. The following is an example edge data file:
InteractionStrength
YAL001C (pp) YBR043W = 0.82
YMR022W (pd) YDL112C = 0.441
YDL112C (pd) YMR022W = 0.9013
Since Cytoscape treats edge data as directional, the second and third edge data values refer to two different edges (source and target are reversed, though the nodes involved are the same).
Each data column is stored in a separate file. Node and edge data files use the same format, and have the suffix ".attrs".
Node and edge data may be loaded via the File → Import → Table menu, just as data table files are.
When expression data is loaded using an expression matrix, it is automatically loaded as node data unless explicitly specified otherwise.
Node and edge data columns are attached to nodes and edges, and so are independent of networks. Data values for a given node or edge will be applied to all copies of that node or edge in all loaded network files, regardless of whether the data file or network file is imported first.
Every data file has one header line that gives the name of the data column, and optionally some additional meta-information about that data column. The format is as follows:
columnName (class=JavaClassName)
The first field is always the column name: it cannot contain spaces. If present, the class field defines the name of the class of the data values. For example, java.lang.String or String for Strings, java.lang.Double or Double for floating point values, java.lang.Integer or Integer for integer values, etc. If the value is actually a list of values, the class should be the type of the objects in the list. If no class is specified in the header line, Cytoscape will attempt to guess the type from the first value. If the first value contains numbers in a floating point format, Cytoscape will assume java.lang.Double; if the first value contains only numbers with no decimal point, Cytoscape will assume java.lang.Integer; otherwise Cytoscape will assume java.lang.String. Note that the first value can lead Cytoscape astray: for example,
floatingPointDataColumn
firstName = 1
secondName = 2.5
In this case, the first value will make Cytoscape think the values should be integers, when in fact they should be floating point numbers. It's safest to explicitly specify the value type to prevent confusion. A better format would be:
floatingPointDataColumn (class=Double)
firstName = 1
secondName = 2.5
or
floatingPointDataColumn
firstName = 1.0
secondName = 2.5
Every line past the first line identifies the name of an object (a node in a node data file or an edge in a edge data file) along with the String representation of the data value. The delimiter is always an equals sign; whitespace (spaces and/or tabs) before and after the equals sign is ignored. This means that your names and values can contain whitespace, but object names cannot contain an equals sign and no guarantees are made concerning leading or trailing whitespace. Object names must be the Node ID or Edge ID as seen in the left-most column of the Table Panel if the data column is to map to anything. These names must be reproduced exactly, including case, or they will not match.
Edge names are all of the form:
sourceName (edgeType) targetName
Specifically, that is
Note that tabs are not allowed in edge names. Tabs can be used to separate the edge name from the "=" delimiter, but not within the edge name itself. Also note that this format is different from the specification of interactions in the SIF file format. To be explicit: a SIF entry for the previous interaction would look like
sourceName edgeType targetName
or
To specify lists of values, use the following syntax:
listDataColumnName (class=java.lang.String)
firstObjectName = (firstValue::secondValue::thirdValue)
secondObjectName = (onlyOneValue)
This example shows a data column whose value is defined as a list of text strings. The first object has three strings, and thus three elements in its list, while the second object has a list with only one element. In the case of a list every data value uses list syntax (i.e. parentheses), and each element is of the same class. Again, the class will be inferred if it is not specified in the header line. Lists are not supported by Styles and so can’t be mapped to network properties.
Sometimes it is desirable to for data values to include linebreaks, such as node labels that extend over two lines. You can accomplish by inserting into the data value. For example:
newlineDataColumn
YJL157C = This is a long\nline for a label.
Is this still relevant?
Formerly, Cytoscape only supported mapping between node/edge IDs and the primary keys in data files. With the introduction of Cytoscape 2.4, this limitation has been removed, and now both IDs and data columns with primitive data types (string, boolean, floating point, and integer) can be selected as the Key Attribute using the drop-down list provided. Complex attributes such as lists are not supported.
Cytoscape uses a simple mechanism to manage aliases of objects. Both nodes and edges can have aliases. If an attribute is loaded as an alias, it is treated as a special attribute called "alias". This will be used when mapping attributes. If the primary key and key attribute for an object do not match, Cytoscape will search for a match between aliases and the key attribute. To define an alias column in the attribute table, just click on the checkboxes to the left of the column name while importing.
When Cytoscape is started, the Table Panel appears in the bottom right of the main Cytoscape window. This browser can be hidden and restored using the F5 key or the View → Show/Hide Table Panel menu option. Like other Panels, the browser can be undocked by pressing the little icon in the top right corner.
To swap between displaying node, edge, and network Data Tables use the
tabs on the bottom of the Table Panel. By default, the Table Panel
displays columns for all nodes and edges in the selected network. To
display columns for only selected nodes/edges, click the Change Table
Mode button
Most column values can be edited by double-clicking the cell (only the ID cannot be edited). Newline characters can be inserted into String columns either by pressing Enter or by typing "\n". Once finished editing, click outside of the editing cell in the Table Panel or press Shift-Enter to save your edits. Pressing Esc while editing will undo any changes.
Rows in the panel can be sorted alphabetically by specific column by
clicking on a column heading. A new column can be created using the
Create New column
It is also possible to import node data columns from public databases via web services, for example from BioMart.
-
Load a network, for example galFiltered.sif.
-
Select File → Import → Table → Public Databases....
-
You will first be asked to select from a set of web services. For this example, we will choose ENSEMBL GENES 73 (SANGER UK).
-
In the Import Data Table from Web Services dialog, select a Data Source. Since galFiltered.sif is a yeast network, select ENSEMBL GENES - SACCHAROMYCES CEREVISIAE.
-
For Key Column in Cytoscape, select shared name for Column and Ensembl Gene ID for Data Type.
The type of identifier selected under Data Type must match what is used in the selected Column in the network.
-
Select the data columns you want to import.
-
Select Import.
When import is complete, you can see the newly imported data columns in the Table Panel.
Annotations in Cytoscape are stored as a set of ontologies (e.g. the Gene Ontology, or GO). An ontology consists of a set of controlled vocabulary terms that annotate the objects. For example, using the Gene Ontology, the Saccharomyces Cerevisiae CDC55 gene has a biological process described as “protein biosynthesis”, to which GO has assigned the number 6412 (a GO ID).
GO 8150 biological_process
-
GO 7582 physiological processes
-
GO 8152 metabolism
-
GO 44238 primary metabolism
-
GO 19538 protein metabolism
- GO 6412 protein biosynthesis
-
-
-
Graphical View of GO Term 6412: protein biosynthesis
Cytoscape can use this ontology DAG (Directed Acyclic Graph) to annotate objects in networks. The Ontology Server (originally called "BioDataServer") is a Cytoscape feature which allows you to load, navigate, and assign annotation terms to nodes and edges in a network. Cytoscape 2.4 now has an enhanced GUI for loading ontology and associated annotation, enabling you to load both local and remote files.
The standard file formats used in the Cytoscape Ontology Server are OBO and Gene Association. The GO website details these file formats:
Cytoscape provides a list of ontologies available in OBO format. If an Internet connection is available, Cytoscape will import ontology and annotation files directly from the remote source. The table below lists the included ontologies.
Ontology Name Description Gene Ontology Full This data source contains a full-size GO DAG, which contains all GO terms. This OBO file is written in version 1.2 format. Generic GO slim A subset of general GO Terms, including higer-level terms only. Yeast GO slim A subset of GO Terms for annotating Yeast data sets maintained by SGD. Molecule role (INOH Protein name/family name ontology) A structured controlled vocabulary of concrete and abstract (generic) protein names. This ontology is a INOH pathway annotation ontology, one of a set of ontologies intended to be used in pathway data annotation to ease data integration. This ontology is used to annotate protein names, protein family names, and generic/concrete protein names in the INOH pathway data. INOH is part of the BioPAX working group. Event (INOH pathway ontology) A structured controlled vocabulary of pathway-centric biological processes. This ontology is a INOH pathway annotation ontology, one of a set of ontologies intended to be used in pathway data annotation to ease data integration. This ontology is used to annotate biological processes, pathways, and sub-pathways in the INOH pathway data. INOH is part of the BioPAX working group. Protein-protein interaction A structured controlled vocabulary for the annotation of experiments concerned with protein-protein interactions. Pathway Ontology The Pathway Ontology is a controlled vocabulary for pathways that provides standard terms for the annotation of gene products. PATO PATO is an ontology of phenotypic qualities, intended for use in a number of applications, primarily phenotype annotation. For more information, please visit the PATO wiki (). Mouse pathology The Mouse Pathology Ontology (MPATH) is an ontology for mutant mouse pathology. This is Version 1. Human disease This ontology is a comprehensive hierarchical controlled vocabulary for human disease representation. For more information, please visit the Disease Ontology website ().
Although Cytoscape can import all kinds of ontologies in OBO format, annotation files are associated with specific ontologies. Therefore, you need to provide the correct ontology-specific annotation file to annotate nodes/edges/networks in Cytoscape. For example, while you can annotate human network data using the GO Full ontology with human Gene Association files, you cannot use a combination of the human Disease Ontology file and human Gene Association files, because the Gene Association file is only compatible with GO.
The Gene Association files provide annotation only for the Gene Ontology. It is a species-specific annotation file for GO terms. Gene Association files will only work with Gene Ontology annotation.
Sample Gene Association File (gene_association.sgd - annotation file for yeast):
SGD S000003916 AAD10 GO:0006081 SGD_REF:S000042151|PMID:10572264 ISS P aryl-alcohol dehydrogenase (putative) YJR155W gene taxon:4932 SGD S000005275 AAD14 GO:0008372 SGD_REF:S000069584 ND C aryl-alcohol dehydrogenase (putative) YNL331C gene taxon:4932
Cytoscape provides a graphical user interface to import both ontology and annotation files at the same time.
Note: All data sources in the preset list are remote URLs, meaning a network connection is required.
- Select File → Import → Ontology and Annotation... to open the "Import Ontology and Annotation" interface. From the Annotation drop-down list, select a gene association file for your network. For example, if you want to annotate the yeast network, select "Gene Association file for Saccharomyces cerevisiae".
* Select an Ontology data (OBO file) from the Ontology drop-down list. If the file is not loaded yet, it will be shown in red. The first three files are Gene Ontology files. You can load other ontologies, but you need your own annotation file to annotate networks.
Once you click the Import button, Cytoscape will start loading OBO and Gene Association files from the remote sources. If you choose GO Full it may take a while since it is a large data file.
- When Cytoscape finishes importing files, the import window will be automatically closed. All columns mapped by this function have the prefix "annotation" and look like this: annotation.[column_name].
Note: Cytoscape supports both OBO formats: version 1.0 and 1.2.
Column data values may be formulas. A typical example is =ABS($otherColumn + LOG(10.2)). Formulas are modeled after Excel™ but only support references to other columns at the same node, edge or network. Since Cytoscape column names may contain embedded spaces, optional braces around the column name (required if the name is not simply a letter followed by one or more letters or digits) is allowed e.g. ${a name with spaces}. Backslashes, opening braces and dollar signs in column names have to be escaped with a leading backslash. For example the column name ex$am{p\le would have to be written as ${ex\$am\{p\\le}. Finally, column names are case sensitive.
String constants are written with double-quotes ". In order to embed a double-quote or a backslash in a string they have to be escaped with a leading backslash, therefore the string "\ must be written as "\"\\". Formula results must be compatible with the type of the column that they have been assigned to. The rules are rather lax though, for example anything can be interpreted as a string and all numeric values will be accepted for a boolean (or logical) column data where non-zero will be interpreted as true and zero as false. For integer columns, floating point values will be converted using the rules of the Excel™ INT function. Parentheses can be used for grouping and to change evaluation order. The operator precedence rules follow those of standard arithmetic.
Currently supported operators are the four basic arithmetic operators
and the ^ exponentiation operator. +, -, *, and
\ are left-associative and ^ is right-associative. The string
concatenation operator is &. Supported boolean or logical operators
are the comparison operators <, >, <=, >=,
=, and <> (not equal).
Currently we support the following functions:
-
Degree -- the degree of a node.
-
InDegree -- the indegree of a node.
-
OutDegree -- the outdegree of a node.
-
SourceID -- the ID of the source node of an edge.
-
TargetID -- the ID of the target of an edge.
-
Abs -- Returns the absolute value of a number.
-
ACos -- Returns the arccosine of a number.
-
ASin -- Returns the arcsine of a number.
-
ATan2 -- Returns the arctangent of two numbers x and y.
-
Average -- Returns the average of a group of numbers.
-
Cos -- Returns the cosine of an angle given in radians.
-
Cosh -- Returns the hyperbolic sine of its argument.
-
Count -- Returns the number of numeric values in a list.
-
Degrees -- Returns its argument converted from radians to degrees.
-
Exp -- Returns e raised to a specified number.
-
Ln -- Returns the natural logarithm of a number.
-
Log -- Returns the logarithm of a number to a specified base.
-
Max -- Returns the maximum of a group of numbers.
-
Median -- Returns the median of a list of numbers.
-
Min -- Returns the minimum of a group of numbers.
-
Mod -- Calculates the modulus of a number.
-
Pi -- Returns an approximation of the value of π.
-
Radians -- Returns its argument converted from degrees to radians.
-
Round -- Rounds a number to a specified number of decimal places.
-
Sin -- Returns the sine of an angle given in radians.
-
Sinh -- Returns the hyperbolic sine of its argument.
-
Sqrt -- Calculates the square root of a number.
-
Tan -- returns the tangent of its argument in radians.
-
Tanh -- returns the hyperbolic tangent of its argument in radians.
-
Trunc -- Truncates a number.
-
Concatenate -- Concatenates two or more pieces of text.
-
Left -- Returns a prefix of s string.
-
Len -- Returns the length of a string.
-
Lower -- Converts a string to lowercase.
-
Mid -- Selects a substring of some text.
-
Right -- Returns a suffix of a string.
-
Substitute -- Replaces some text with other text.
-
Text -- Format a number using the Java DecimalFormat class' conventions.
-
Upper -- Converts a string to uppercase.
-
Value -- Converts a string to a number.
-
And -- Returns the logical conjunction of any number of boolean values.
-
Not -- Returns the logical negation of a boolean value.
-
Or -- Returns the logical disjunction of any number of boolean values.
-
First -- Returns the first entry in a list.
-
Last -- Returns the last entry in a list.
-
Nth -- Returns the n-th entry in a list.
-
Largest -- the kth largest value in a list.
-
GeoMean -- the geometric mean of a set of numbers.
-
HarMean -- the harmonic mean of a set of numbers.
-
Mode -- the mode of a set of numbers.
-
NormDist -- Returns the pdf or CDF of the normal distribution.
-
Permut -- Returns the number of permutations for a given number of objects.
-
StDev - sample standard deviation.
-
Var -- sample variance.
-
Combin - Returns the number of combinations for a given number of objects.
-
If -- Returns one of two alternatives based on a boolean value.
-
ListToString -- Returns a string representation of a list.
-
Now -- Returns a string representation of the current date and time.
-
Today -- returns a string representation of the current date.
The possibly biggest problem is the referencing of other columns that have null values. This is not allowed and leads to errors. In order to mitigate this problem we support the following optional syntax for column references: ${columnName:defaultValue}. The interpretation is that if columnName is null, then the default value will be used, otherwise the value of the referenced value will be used instead. The referenced column must still be a defined column and not an arbitrary name! The other potential problem is when there are circular column reference dependencies. Circular dependencies will be detected at formula evaluation time and lead to a run-time error.
When working with formulas it can be very helpful to open the Developer's Log Console. Formula evaluation errors will be logged there.
In order to ease the creation of formulas as well as to facilitate
discovery of built-in functions we provide a Function Builder in the
Table Panel. After selecting a non-list column cell, you can invoke it
by clicking on
The Function Builder is a useful tool for discovery of the list of built-in functions, which has the return type matching the data type of the column. Arguments can either be selected from a list of named columns or constant values can be entered in a text entry field. A major shortcoming at this time is that the Formula Builder won't let you compose functions with function calls as arguments. If you need the most general functionality, please type the expression directly into a cell.
It is relatively easy to add your own built-in formula functions. A simple function can probably be implemented in 15 to 20 minutes. It can then be registered via the parser and becomes immediately available to the user. It will of course also show up in the drop-down list in the Function Builder.
You can search for nodes and edges by column value directly through
Cytoscape's tool bar. For example, to select nodes or edges with a
column value that starts with "STE", type ste* in the search bar. The
search is case-insensitive. The * is a wildcard character that matches
zero or more characters, while "?" matches exactly one character. So
ste? would match "STE2" but would not match "STE12". Searching for
ste* would match both.
To search a specific column, you can prefix your search term with the
column name followed by a :. For example, to select nodes and edges
that have a "COMMON" column value that starts with "STE", use
common:ste*. If you don't specify a particular column, all columns
will be searched.
Columns with names that contain spaces, quotes, or characters other than letters and numbers currently do not work when searching a specific column. This will be fixed in a future release.
To search for column values that contain special characters you need to
escape those characters using a "\". For example, to search for
"GO:1232", use the query GO\:1232. The complete list of special
characters is:
+ - & | ! ( ) { } [ ] ^ " ~ * ? : \
Note: Escaping characters only works when searching all columns. It currently does not work for column-specific searching. This will be fixed in a future release.
Cytoscape 3 provides a new user interface for filtering nodes and edges. These tools can be found in the Select panel:
There are two main types of filters: narrowing filters, which can be combined into a tree; and chainable filters, which can only be combined in a linear chain. These can be found in the Filter, and Chain tabs of the Select panel, respectively.
Narrowing filters select nodes and edges based on user-specified constraints. For example, you can use these filters to find edges with a weight between 0 and 5.5, or nodes with degree less than 3. A filter can contain an arbitrary number of sub-filters. To add one to your filter, click on the "+" button.
Due to the nature of narrowing filters, Cytoscape can apply them to a network efficiently and interactively. Some filters even provide slider controls to quickly explore different thresholds. This is the default behavior on smaller networks. For larger networks, Cytoscape automatically disables this interactivity. You can override this by manually checking the Apply Automatically box above the Apply button:
Cytoscape comes packaged with the following narrowing filters:
This filter will match nodes or edges that have particular column values. For numeric columns, sliders are provided to set thresholds in interactive mode. Otherwise, minimum and maximum values can be keyed in.
From string columns, a variety of matching options are provided:
For example, column values can be checked if they contain or match exactly a particular string. The filter can also match against Java-style regular expressions. A checkbox controls whether the matching is done in a case-sensitive way.
The degree filter matches nodes with a degree that falls within the given minimum and maximum values, inclusive. You can choose whether the filter operates on the in-degree, out-degree or overall (in + out) degree.
The topology filter matches nodes with a certain number of neighbors which are within a fixed distance away. The thresholds for the neighborhood size and distance can be set independently.
By default, nodes and edges need to satisfy the constraints of all your
filters. You can change this so that instead, only the constraints of at
least one filter needs to be met in order to match a node or edge. This
behavior is controlled by the Match all/any drop-down box. This
appears once your filter has more than one sub-filter. For example,
suppose you wanted to match nodes with column COMMON containing ste
or cdc, but you only want nodes with degree 5 or more, you'd first
construct a filter that looks like this:
This filter will match nodes where COMMON contains ste and
cdc. To change this to a logical or operation, drag either of the
column filters by its handle onto the other column filter to create a
new group. Now change the group's matching behavior to Match any:
You can also reorder filters by dropping them in-between existing filters.
Chainable filters are a more general case of narrowing filters. These filters take nodes and edges, from your selection or a narrowing filter, as input. The nodes and edges in the output of the filter become the input of the next filter in the chain. The output of the last filter becomes the new selection.
Chainable filters can also be reordered by dragging one by the handle and dropping it between existing filters.
Cytoscape currently bundles the following chainable filter:
This transformer will go through all the input edges and add their source nodes, target nodes, or both, to the output. This is useful for selecting nodes that are connected to edges that match a particular filter.
The name of active filter appears in the drop-down box at the top of Select panel. Beside this is the options button which will allow you to rename, remove or export the active filter. It also lets you create a new filter, or import filters.
At the bottom of the Select panel, there is an Apply button that will re-apply the active filter. On the opposite side of the progress bar is the cancel button, which will let you interrupt a long-running filter.
The Select → Nodes and Select → Edges menus provide several mechanisms for selecting nodes and edges. Most options are fairly straightforward; however, some need extra explanation.
Select → Nodes → From ID List File... selects nodes based on node identifiers found in a specified file. The file format is simply one node id per line:
Node1
Node2
Node3
...
Cytoscape uses a Zoomable User Interface for navigating and viewing networks. ZUIs use two mechanisms for navigation: zooming and panning. Zooming increases or decreases the magnification of a view based on how much or how little a user wants to see. Panning allows users to move the focus of a screen to different parts of a view.
Cytoscape provides two mechanisms for zooming: toolbar buttons and the scroll wheel. Use the zooming buttons located on the toolbar to zoom in and out of the interaction network shown in the current network display. Zoom icons are detailed below:
From Left to Right:
-
Zoom In
-
Zoom Out
-
Zoom Out to Display all of Current Network
-
Zoom Selected Region
Using the scroll wheel, you can zoom in by scrolling up and zoom out by scrolling downwards. These directions are reversed on Macs with natural scrolling enabled (the default for Mac OS X Lion and newer versions).
There are several ways to pan the network:
-
Middle Click and Drag - You can pan the network image by holding down the middle mouse button and moving the mouse.
-
Command Key + Drag (Mac only) - If you use Mac without middle button, you can pan the view by holding down Command key and then drag.
-
Dragging Box on Network Overview - You can also pan the image by holding down the left mouse button over the blue box in the overview panel in the lower part of the Network tab in the Control Panel.
Click the left mouse button on a node or edge to select that object. You can hold down the Shift key (or Ctrl key on PCs) to select more than one node/edge or you can hold down the left mouse button and drag the mouse to select groups of nodes/edges.
Click the right mouse button (or Ctrl+left mouse button on Macs) on a node/edge to launch a context-sensitive menu with additional information about the node/edge.
This menu can change based on the current context. For nodes, it typically shows:
-
Add
-
Edit
-
Select
-
Group
-
Nested Networks
-
Apps
-
External Links
-
Preferences
Edges usually have the following menu:
-
Edit
-
Select
-
Apps
-
External Links
-
Preferences
Apps can contribute their own items into node and edge context menus. These additions usually appear in the Apps section of the context menu.
-
Add Nested Network: Lets the user select any network in Cytoscape as the current node's nested network. If the current node already has a nested network it will be replaced.
-
Remove Nested Network: Removes the currently associated nested network from a node. The associated network is not deleted. Only the association between the node and the network is removed.
-
Go to Nested Network: The current node's nested network will be the current network view and have the focus. Should a network view for the nested network not exist, it will be created.
More information about nested networks can be found in the Nested Networks section.
The Layout menu has an array of features for organizing the network visually according to one of several algorithms, aligning and rotating groups of nodes, and adjusting the size of the network. Cytoscape layouts have three different sources, which are reflected in the Layout menu.
With the exception of the yFiles layouts (explained below), Cytoscape Layouts have the option to operate on only the selected nodes, and all provide a Settings... panel to change the parameters of the algorithm. Most of the Cytoscape layouts also partition the graph before performing the layout. In addition, many of these layouts include the option to take either node or edge columns into account. A few of the layout algorithms are:
The grid layout is a simple layout the arranges all of the nodes in a square grid. This is the default layout and is always available as part of the Cytoscape core. It is available by selecting Layout → Grid Layout. A sample screen shot is shown above.
The spring-embedded layout is based on a “force-directed” paradigm as implemented by Kamada and Kawai (1988). Network nodes are treated like physical objects that repel each other, such as electrons. The connections between nodes are treated like metal springs attached to the pair of nodes. These springs repel or attract their end points according to a force function. The layout algorithm sets the positions of the nodes in a way that minimizes the sum of forces in the network. This algorithm can be applied to the entire network or a portion of it by selecting the appropriate options from Layout → Edge-weighted Spring Embedded.
The Attribute Circle layout is a quick, useful layout, particularly for small networks, that will locate all of the nodes in the network around a circle. The node order is determined by a user-selected node column. The result is that all nodes with the same value for that column are located together around the circle. Using Layout → Attribute Circle Layout → column to put all nodes around a circle using column to position them. The sample screen shot above shows the a subset of the galFiltered network organized by node degree.
The Group Attributes layout is similar to the Attribute Circle layout described above except that instead of a single circle with all of the nodes, each set of nodes that share the same value for the column are laid out in a separate circle. The same network shown above (network generated by PSICQUIC Client) is shown above, using Layout → Group Attributes Layout → taxonomy.
The force-directed layout is a layout based on the "force-directed" paradigm. This layout is based on the algorithm implemented as part of the prefuse toolkit provided by Jeff Heer. The algorithm is very fast and with the right parameters can provide a very visually pleasing layout. The Force Directed Layout will also accept a numeric edge column to use as a weight for the length of the spring, although this will often require more use of the Settings... dialog to achieve the best layout. This algorithm is available by selecting Layout → Prefuse Force-Directed Layout → (unweighted) or the edge column you want to use as a weight. A sample screen shot showing a portion of the galFiltered network provided in sample data is provided above.
yFiles layouts are a set of commercial layout algorithms which are provided courtesy of yWorks. Due to license restrictions, the detailed parameters for these layouts are not available (there are no yFiles entries in the Layout → Settings...). The main layout algorithms provided by yFiles are:
The organic layout algorithm is a kind of spring-embedded algorithm that combines elements of the other algorithms to show the clustered structure of a graph. This algorithm is available by selecting Layout → yFiles Layouts → Organic.
This algorithm produces layouts that emphasize group and tree structures within a network. It partitions the network by analyzing its connectivity structure, and arranges the partitions as separate circles. The circles themselves are arranged in a radial tree layout fashion. This algorithm is available by selecting Layout → yFiles Layouts → Circular.
The hierarchical layout algorithm is good for representing main direction or "flow" within a network. Nodes are placed in hierarchically arranged layers and the ordering of the nodes within each layer is chosen in such a way that minimizes the number of edge crossings. This algorithm is available by selecting Layout → yFiles Layouts → Hierarchical.
Many layouts have adjustable parameters that are exposed through the Layouts → Settings... menu option. The Layout Settings dialog, which allows you to choose which layout algorithm settings to adjust, is shown below. The settings presented vary by algorithm and only those algorithms that allow access to their parameters will appear in the drop-down menu at the top of the dialog. Once you've modified a parameter, clicking the Execute Layout button will apply the layout.
From Cytoscape 3.0, Edge Bend is a regular edge property and can be used as a part of a Style. Just like any other edge property, you can select a Default Value, a Mapping and use Bypass for select nodes. In the Styles tab, select the Bend property from the Properties drop-down and click on either the Default Value, Mapping or Bypass cell to bring up the Edge Bend Editor. In the editor, you can add as many handles as you want to the edge using Alt-Click on Windows, Option-Click on Mac, or Ctrl-Alt-Click on Linux.
To clear all edge bends, select Layout → Clear All Edge Bends.
In addition to adding handles manually, you can use the Bundle Edges function to bundle all or selected edges automatically.
-
Select Layout → Bundle Edges → All Nodes and Edges.
-
Set parameters.
- Details of the algorithm is described in this paper.
-
Press OK to run. Edge bundling may take a long time if the number of edges is large.
-
If it takes too long, try decreasing Maximum Iterations.
-
For large, dense networks, try setting Maximum iterations in the range of 500 - 1000.
-
Note: The handle locations will be optimized for current location of nodes. If you move node positions, you need to run the function again to get proper result.
The simplest method to manually organize a network is to click on a node and drag it. If you select multiple nodes, all of the selected nodes will be moved together.
Selecting the Layout → Rotate option will show the Rotate window in the Tool Panel. This function will either rotate the entire network or a selected portion of the network. The image below shows a network with selected nodes rotated.
Before
After
Selecting the Layout → Scale option will open the Scale window in the Tool Panel. This function will scale the position of the entire network or of the selected portion of the network. Note that only the position of the nodes will be scaled, not the node sizes. Node size can be adjusted using Styles. The image below shows selected nodes scaled.
Before
After
Selecting the Layout → Align/Distribute option will open the Align and Distribute window in the Tool Panel. Align provides different options for either vertically or horizontally aligning selected nodes against a line. The differences are in what part of the node gets aligned, e.g. the center of the node, the top of the node, the left side of the node. Distribute evenly distributes selected nodes between the two most distant nodes along either the vertical or horizontal axis. The differences are again a function what part of the node is used as a reference point for the distribution. Stack vertically or horizontally stacks selected nodes with the full complement of alignment options. The table below provides a description of what each button does.
Button Before After Description of Align Options
Button Before After Description of Distribute Options
Button Before After Description of Stack Options
In addition to the ability to click on a node and drag it to a new position, Cytoscape now has the ability to move nodes using the arrow keys on the keyboard. By selecting one or more nodes using the mouse and clicking one of the arrow keys (←, →, ↑, ↓) the selected nodes will move one pixel in the chosen direction. If an arrow key is pressed while holding the Shift key down, the selected nodes will 15 pixels in the chosen direction.
One of Cytoscape's strengths in network visualization is the ability to allow users to encode any table data (name, type, degree, weight, expression data, etc.) as a property (such as color, size of node, transparency, or font type) of the network. A set of these encoded or mapped table data sets is called a Style and can be created or edited in the Style panel of the Control Panel. In this interface, the appearance of your network is easily customized. For example, you can:
Specify a default color and shape for all nodes.
Set node sizes based on the degree of connectivity of the nodes. You can visually see the hub of a network...
...or, set the font size of the node labels instead.
Visualize gene expression data along a color gradient.
Encode specific physical entities as different node shapes.
Use specific line types to indicate different types of interactions.
Control edge transparency (opacity) using edge weights.
Control multiple edge properties using edge score.
Browse extremely-dense networks by controlling the opacity of nodes.
Show highly-connected region by edge bundling and opacity.
Add photo/image/graphics on top of nodes.
Cytoscape 3 has several sample styles. Below are a few examples of these applied to the galFiltered.sif network :
The Style interface is located under the Style panel of the Control Panel.
This interface allows you to create/delete/view/switch between different styles using the Current Style options. The panel displays the mapping details for a given style and is used to edit these details as well.
-
At the top of the interface, there is a drop-down menu for selecting a pre-defined style. There is also an Options drop-down with options to rename, remove, create and copy a Style, and an option to create a legend for the selected Style.
-
The main area of the interface is composed of three tabs, for Node, Edge and Network.
-
Each tab contains a list of properties relevant to the current style. At the top of the list a Properties drop-down allows you to add additional properties to the list.
-
Each property entry in the list has 3 columns:
-
The Default Value shows just that, the default value for the property. Clicking on the Default Value column for any property allows you to change the default value.
-
Mapping displays the type of mapping currently in use for the property. Clicking on the Mapping column for any property expands the property entry to show the interface for editing the mapping. Details on the mapping types provided here.
-
Bypass displays any style bypass for a selected node or edge. Note that a node/edge or subset of nodes/edges must be selected to activate the Bypass column. Clicking on the Bypass column for selected node(s)/edge(s) allows you to enter a bypass for that property for selected node(s)/edge(s).
-
The Default Value is used when no mapping is defined for a property, or for nodes/edges not covered by a mapping for a particular property. If a Mapping is defined for a property, this defines the style for all or a subset of nodes/edges, depending on how the mapping is defined. A Bypass on a specific set of nodes/edges will bypass and override both the default value and defined mapping.
The Cytoscape distribution includes several predefined styles to get you started. To examine a few styles, try out the following example:
Step 1. Load some sample data
-
Load a sample session file: From the main menu, select File → Open, and select the file sampleData/galFiltered.cys.
-
The session file includes a network, some annotations, and sample styles. By default, the Sample for galFiltered style is selected. Gene expression values for each node are colored along a color gradient between red and green (where red represents a low expression ratio and green represents a high expression ratio, using thresholds set for the gal4RGexp experiment bundled with Cytoscape in the sampleData/galExpData.csv file). Also, node size is mapped to the degree of the node (number of edges connected to the node) and you can see the hubs of the network as larger nodes. See the sample screenshot below:
-
Step 2. Switch between different styles
You can change the style by making a selection from the Current Style drop-down list, found at the top of the Style panel.
For example, if you select Sample1, a new style will be applied to your network, and you will see a white background and round blue nodes. If you zoom in closer, you can see that protein-DNA interactions (specified with the label "pd") are drawn with dashed edges, whereas protein-protein interactions (specified with the label "pp") are drawn with solid edges (see sample screenshot below).
Finally, if you select Solid, you can see the graphics below:
This style does not have mappings except node/edge labels, but you can modify the network graphics by editing the Default Value for any property.
Additional sample styles are available in the sampleStyles.xml file in
the sampleData directory. You can import the sample file from File →
Import → Style File....
Cytoscape allows a wide variety of properties to be controlled. These are summarized in the tables below.
Node Properties Description Border Line Type The type of line used for the border of the node. Border Transparency The opacity of the color of the border of the node. Allows for transparency. Border Width The width of the node border. Label The text used for the node label. Label Font Face The font used for the node label. Label Font Size The size of the font used for the node label. Label Position The position of the node label relative to the node. Label Transparency The transparency of the color of the node label. Allows for transparency. Label Width The maximum width of the node label. If the node label is wider than the specified width, Cytoscape will automatically wrap the label on space characters. Cytoscape will not hyphenate words, meaning that if a single word (i.e. no spaces) is longer than maximum width, the word will be displayed beyond the maximum width. Nested Network Image Visible A boolean value that indicates whether a nested network should be visualized (assuming a nested network is present for the specified node). Paint > Paint The color of the node. Paint > Border Paint The color of the border of the node. Paint > Custom Graphics 1-9 A user-defined custom graphic that is displayed on the node. Paint > Custom Graphics Position 1-9 The position of each custom graphic. Paint > Fill Color The color of the node. Paint > Label Color The color of the node label. Paint > Selected Paint The color of the node when selected. Shape The shape of the node. Size > Size The size of the node. Width and height will be equal. This property is mutually exclusive of Node Height and Node Width. Size > Custom Graphics Size 1-9 A user-defined custom graphic size that is displayed on the node. Size > Height The height of the node. Height will be independent of width. This property is mutually exclusive of Node Size. Size > Width The width of the node. Width will be independent of height. This property is mutually exclusive of Node Size. Size > Fit Custom Graphics to node Toggle to fit Custom Graphics size to node size. Size > Lock node width and height Toggle to ignore Width and Height, and to use Size for both values. Tooltip The text of the tooltip that appears when a mouse hovers over the node. Transparency The opacity of the color of the node. Zero means totally transparent, and 255 is most opaque. Visible Node is visible or not. By default, this value is set to true. X Location X location of the node. Default value of this will be ignored. The value will be used only when mapping function is defined. Y Location Y location of the node. Default value of this will be ignored. The value will be used only when mapping function is defined. Z Location Z location of the node. Default value of this will be ignored. The value will be used only when mapping function is defined.
Edge Properties Description Bend The edge bend. Defines how the edge is rendered. Users can add multiple handles to define how to bend the edge line. Curved If Egde Bend is defined, edges will be rendered as straight or curved lines. If this value is set to true, edges will be drawn as curved lines. Label The text used for the edge label. Label Font Face The font used for the edge label. Label Font Size The size of the font used for the edge label. Label Transparency The opacity of the color of the edge label. Allows for transparency. Line Type The type of stoke used to render the line (solid, dashed, etc.) Paint > Paint The color of the edge. Paint > Color (Selected) > Color (Selected) The color of the edge. Paint > Color (Selected) > Source Arrow Selected Paint The selected color of the arrow on the source node end of the edge. Paint > Color (Selected) > Stroke Color (Selected) The color of the edge when selected. Paint > Color (Selected) > Target Arrow Selected Paint The selected color of the arrow on the target node end of the edge. Paint > Color (Unselected) > Color (Unselected) The color of the edge. Paint > Color (Unselected) > Source Arrow Unselected Paint The color of the arrow on the source node end of the edge. Paint > Color (Unselected) > Stroke Color (Unselected) The color of the edge. Paint > Color (Unselected) > Target Arrow Unselected Paint The color of the arrow on the target node end of the edge. Paint > Label Color The color of the edge label. Source Arrow Shape The shape of the arrow on the source node end of the edge. Target Arrow Shape The shape of the arrow on the target node end of the edge. Tooltip The text of the tooltip that appears when a mouse hovers over the edge. Transparency The opacity of the color of the edge. Allows for transparency. Visible Edge is visible or not. By default, this value is set to true. Width The width of the line.
Network Properties Description Background Paint The background color of the network view. Center X Location The X location of network view center. Center Y Location The Y location of network view center. Edge Selection Edges are selectable or not. If this is false, users cannot select edges. Node Selection Nodes are selectable or not. If this is false, users cannot select nodes. Scale Factor The zoom level of the network view. Size > Size The size of the network view. Size > Height The height of the network view. Size > Width The width of the network view. Title The title of the network view.
Available Shapes and Line Styles Sample
Node Shapes 
For each property, you can specify a default value or define a dynamic mapping. Cytoscape currently supports three different types of mappings:
-
Passthrough Mapping
- The values of network column data are passed directly through to properties. A passthrough mapping is typically used to specify node/edge labels. For example, a passthrough mapping can label all nodes with their common gene names.
-
Discrete Mapping
- Discrete column data are mapped to discrete properties. For example, a discrete mapping can map different types of molecules to different node shapes, such as rectangles for gene products and ellipses for metabolites.
-
Continuous Mapping
-
Continuous data are mapped to properties. Depending on the property, there are three kinds of continuous mapping:
i. Continuous-to-Continuous Mapping: for example, you can map a continuous numerical value to node size.
ii. Color Gradient Mapping: This is a special case of continuous-to-continuous mapping. Continuous numerical values are mapped to a color gradient.
iii. Continuous-to-Discrete Mapping: for example, all values below 0 are mapped to square nodes, and all values above 0 are mapped to circular nodes.
-
However, note that there is no way to smoothly morph between circular nodes and square nodes.
-
The table below shows mapping support for each property.
Legend
Symbol Description
-
Mapping is not supported for the specified property.
-
Mapping is fully supported for the specified property.
o Mapping is partially supported for the specified property. Support for “continuous to continuous” mapping is not supported.
Node Mappings
Node Property Passthrough Mapping Discrete Mapping Continuous Mapping Color Fill Color + + + Transparency + + + Border Paint + + + Border Transparency + + + Label Color + + + Label Transparency + + + Numeric Size/Width/Height + + + Label Font Size + + + Border Width + + + Label Width + + + Other Border Line Type + + o Shape + + o Label + + o Tooltip + + o Label Font Face + + o Label Position o + o Nested Network Image Visible + + o
Edge Mappings
Edge Properties Passthrough Mapping Discrete Mapping Continuous Mapping Color Color + + + Transparency + + + Target Arrow Color + + + Source Arrow Color + + + Label Color + + + Label Transparency + + + Numeric Width + + + Label Font Size + + + Label Width + + + Other Line Type + + o Source Arrow Shape + + o Target Arrow Shape + + o Label + + o Tooltip + + o Label Font Face o + o
In Cytoscape 2.8.0 and later versions, the Passthrough Mapping can recognize some text representations of values. This means, if you have a string column named Node Size Values, you can directly map those values as the Node Size by setting Node Size Values as controlling column with Node Size Passthrough mapping. The following value types are supported:
-
Color: Standard color names supported by all browsers or RGB representation in hex
-
Numerical Values: Automatically mapped to the specified property.
-
Custom Graphics: URL String. If the URL is valid and an actual image data exists there, Cytoscape automatically downloads the image and maps it to the node.
-
Color Passthrough Mapping
-
Node Size Passthrough Mapping
-
Custom Graphics Passthrough Mapping
For Cytoscape 2.8.0 and later versions, Cytoscape supports Custom Graphics for nodes. Using the Style interface, you can map Custom Graphics to nodes like any other property. Cytoscape has a set of graphics and you can also add your own graphics in the Custom Graphics Manager, as well as remove or modify existing graphics.
Taxonomy Icon set used in this section is created by Database Center for Life Science (DBCLS) and is distributed under Creative Commons License (CC BY 2.1.)
The Custom Graphics Manager is available under View → Open Custom Graphics Manager:
-
-
You can add images by drag-and-drop of image files and URLs. If you want to add images from a web browser or local file system, you can drag images from them and drop those images onto the list of images on the left.
- Note: When you drag and drop images from web browser, make sure that you are actually dragging the URL for the image. In some cases, images are linked to an HTML page or scripts, and in such cases, this drag and drop feature may not work.
-
If you want to add all images in a folder, press the + button on the bottom of the Custom Graphics Manager window.
-
-
To remove images from the current session's Custom Graphics library, simply select graphics from the list and press the - button.
-
Images can be resized by defining specific Width and Height information. If the Aspect Ratio box is checked, the width-height ratio is always synchronized. You can resize the image to the original size by pressing Original button.
Custom Graphics is used and defined like any other property, through the Style interface. There are nine Custom Graphics properties.
-
To select a custom graphic, first add the Custom Graphics property to the Properties list in the Style interface (on the Node tab, select Properties → Paint → Custom Paint n → Custom Graphics n). Next, click the Default Value column of the Custom Graphics property to bring up the Graphics dialog. Select an image, a chart or a gradient and then click Apply.
- By default, custom graphics objects are automatically resized to be consistent with the Node Size property.
-
To remove a custom graphics, click the Remove Graphics button on the Graphics dialog.
-
-
Cytoscape provides three kinds of Custom Graphics:
-
Images: You can select one of the provided images or add your own (use the Custom Graphics Manager to add more images to the list).
-
Charts: The following chart types are available:
Bar ,
Box,
Heat Map,
Line,
Pie,
Ring.
-
Gradients: You can also set Linear and Radial gradients to nodes.
-
Each custom graphics property is associated with a position. You can edit their positions by using the UI available in the Default Value column for any Custom Graphics Position property.
- Note: Setting custom graphics positions for Linear or Radial gradients has no effect, as they are always centered on the node.
The number that appears with the custom graphics property represents an ordering of layers. Basic node color and shape are always rendered first, then node custom graphics 1, 2, ..., through 9.
In general, saving and loading Custom Graphics is automatic. When you quit Cytoscape, all of the Custom Graphics in the manager will be saved automatically. There are two types of saving:
-
To session file
- When you save the current session to a file, the Custom Graphics used in Style will be saved to that file. For example, if you have a style with a discrete mapping for Custom Graphics, all custom graphics used in the style will be saved to the session file. Other graphics will not be saved in your session file. This is because your image library can be huge when you add thousands of images to the Custom Graphics Manager and it takes very long time to save and load the session file.
-
Automatic saving to
CytoscapeConfiguration/images3directory- When you select File → Quit, all of Custom Graphics in the
Manager will be saved automatically to your Cytoscape
setting directory. Usually, it's
YOUR_HOME_DIRECTORY/CytoscapeConfiguration/images3.
- When you select File → Quit, all of Custom Graphics in the
Manager will be saved automatically to your Cytoscape
setting directory. Usually, it's
In any case, Custom Graphics will be saved automatically to your system or session and will be restored when you restart Cytoscape or load a session.
The following tutorials demonstrate some of the basic Style features. Each tutorial is independent of the others.
The goal of this tutorial is to learn how to create a new Style and set some default values.
-
Load a sample network. From the main menu, select File → Import → Network → File..., and select
sampleData/galFiltered.sif. -
Create some node/edge statistics by Network Analyzer. Network Analyzer calculates some basic statistics for nodes and edges. From the main menu, select Tools → Network Analyzer → Network Analysis → Analyze Network, and click OK. Once the result is displayed, simply close the window. All statistics are stored as regular table data.
-
Select the Style panel in the Control Panel.
-
Create a new style. Click the Options
drop-down, and select Create New Style. Enter a name for your new style when prompted.
Since no mappings are set up yet, only default values are defined for some of the properties. From this panel, you can create node/edge mappings for all properties.
-
Change the default node color and shape. To set the default node shape to triangles, click the Default Value column for Node Shape. A list of available node shapes will be shown. Select the Round Rectangle icon and click the Apply button. You can edit other default values in the same way. In the example shown below, the node shape is set to Round Rectangle, while the Node Fill Color is set to white. The new Style is automatically applied to the current network, as shown below.
Now you have a network with new Style. The following section demonstrates how to create a new style using a discrete mapping. The goal is to draw protein-DNA interactions as dashed lines, and protein-protein interactions as solid lines.
-
Choose a property. In the Edge tab of the Style panel, find the Stroke Color (Unselected) property.
-
Choose a data column to map to. Expand the entry for Stroke Color (Unselected) by clicking the arrow icon to the right. Click the Column entry and select "interaction" from the drop-down list that appears.
-
Choose a mapping type. Under Mapping Type, select Discrete Mapping. All available column values for "interaction" will be displayed, as shown below.
-
Set the mapping relationship. Click the empty cell next to "pd" (protein-DNA interactions). On the right side of the cell, click on the ... button that appears. A popup window will appear; select green or similar, and the change will immediately appear on the network window.
Repeat step 4 for "pp" (protein-protein interactions), but select a darker color as edge stroke color. Then repeat steps 3 through 4 for the Edge Line Type property, by selecting the correct line style ("Dash" or "Solid") from the list.
Now your network should show "pd" interactions as dashed green lines and "pp" interactions as solid lines. A sample screenshot is provided below.
At this point, you have a network with some edge mappings. Next, let's create mappings for nodes. The following section demonstrates how to create a new style using a continuous mapping. The goal is to superimpose node statistics (in this example, node degree) onto a network and display it along a color gradient.
-
Choose a property. In the Node tab of the Style panel, find the Fill Color property.
-
Choose a node table column. Expand the entry for Fill Color by clicking the arrow icon to the right. Click the Column entry and select "Degree" from the drop-down list that appears.
-
Choose a mapping type. Set the "Continuous Mapping" option as the Mapping Type. This automatically creates a default mapping.
-
Define the points where colors will change. Double-click on the black-and-white gradient rectangle next to Current Mapping to open the Color Gradient Mapping. Note the two smaller triangles at the top of the gradient. Move the left-side one all the way to the left, and the right-side one all the way to the right.
-
Define the colors between points. Double-click on the larger leftmost triangle (facing left) and a color palette will appear. Set the color white and repeat for the smaller left-side triangle. For the triangle at the right, set its color to green and then choose the same green color for the smaller right-side triangle.
The color gradients will immediately appear on the network. All nodes with degree 1 will be set to white, and all values between 1 and 18 will be painted with a white/green color gradient. A sample screenshot is below.
-
-
Repeat for other properties. You can create some more mappings for other numeric table data. For example, edge data table column "EdgeBetweenness" is a number, so you can use it for continuous mapping. The following is an example visualization by mapping Edge Width to "EdgeBetweenness".
The following tutorial demonstrates utilities for editing discrete mappings. The goal of this section is learning how to set and adjust values for discrete mappings automatically.
-
Switch the Current Style to Minimal. Now your network looks like the following:
-
Create a discrete Node Color mapping. Select "AverageShortestPathLength" (generated by Network Analyzer) as the controlling property.
-
Click the Fill Color cell, then right-click it and select Mapping Value Generators → Rainbow. Cytoscape will automatically generate different colors for different property values as shown below:
-
Create a discrete Node Label Font Size mapping. Select "AverageShortestPathLength" as controlling property.
-
Click the Node Label Font Size cell, then right-click it and select Mapping Value Generators → Number Series. Type 3 for the first value and click OK. Enter 3 for increment.
-
Apply Layout → yFiles Layouts → Organic. The final view is shown below:
This mapping generator utility is useful for categorical data. The following example is created by adding node color discrete mapping from species column to color.
This tutorial is a quick introduction to Custom Graphics feature. You can assign up to nine images per node as a part of a Style.
-
In this first example, we will use the presets that Cytoscape 3 has. In general, you can use any type of bitmap graphics. Custom images should be added to the Custom Graphics Manager before using them in a Style.
-
If you are continuing from the previous tutorials, skip to the next step. Otherwise, load a network and run the Network Analyzer (Tools → Network Analyzer → Network Analysis → Analyze Network). This creates several new table columns (statistics for nodes and edges).
-
Click the Style panel in the Control Panel, and select the Solid style.
-
Under Properties, add Custom Graphics 1 from Paint → Custom Paint 1 → Custom Graphics 1.
-
In the Default Value cell of the Custom Graphics 1 entry, click to open the Graphics dialog, and select any of the custom graphics from the list.
-
In the Default Value cell of Transparency, set the value to zero.
-
Set the Default Value of node Size to 80. Also change the edge width to 6 and the node label font size to 10, to increase readability.
-
Press Apply. Now your network looks like the following:
-
Open the Custom Graphics Manager under View → Open Custom Graphics Manager. Drag and Drop this
icon to the image list which automatically adds it to the manager.
-
Create a continuous mapping for Node Custom Graphics 2. Select "BetweennessCentrality" as controlling property. In the Continuos Mapping Editor, add a handle position by clicking in the Add button, and move the handle to 0.2. Double-click the region over 0.2 and set the new icon you have just added in the last step.
-
Press Apply. Under Properties, add Custom Graphics Position 2 from Paint → Custom Paint 2 → Custom Graphics Position 2. Change the Default Value and move the position of the graphics to upper left.
-
Press Apply. Now the important nodes in the network (nodes with high betweenness centrality) are annotated with the icon.
The goal of this tutorial is learning how to create and customize node charts from data stored in the Node tables.
-
Start a new session and load a sample network. From the main menu, select File → Import → Network → File..., and select
sampleData/galFiltered.sif. -
Create some node/edge statistics by Network Analyzer. Network Analyzer calculates some basic statistics for nodes and edges. From the main menu, select Tools → Network Analyzer → Network Analysis → Analyze Network, and click OK. Once the result is displayed, simply close the window. All statistics are stored as regular table data.
-
Select the Style panel in the Control Panel.
-
Create a new style. Click the Options
-
Add Custom Graphics 1 to the properties list. Click Properties and select Paint → Custom Paint 1 → Custom Graphics 1.
-
In the Default Value cell of the Custom Graphics 1 property entry, click to open the Graphics dialog.
-
Click the Charts tab and make sure the Bar Chart option is selected.
-
Select data columns. Now you have to choose the columns in the Node table that contains the data you want to be displayed as charts. The Available Columns list displays all node columns that can be used as chart data (i.e. single or list columns of numerical types).
-
First click the Remove All button to remove the current selected columns (by default, Cytoscape selects the first column in the Available Columns list).
-
Now select all centrality and coefficient columns from Available Columns list and click the Add Selected button.
-
-
Click the Apply button to create bar charts with the selected data columns and default options.
-
The network view doesn't look so good yet, so let's make a few changes to its Style before we continue. In the example shown below, the node Shape is set to Rectangle, while the node Fill Color is set to white. It's also a good idea to apply another layout (e.g. select Layout → yFiles Layouts → Organic).
-
Focus on one node to see the chart details. For example search for and then focus on node "YMR043W".
-
Change other chart options. Click the Default Value cell of the Custom Graphics 1 property again in order to open the Graphics dialog and then select the Options tab on the Bar Chart editor.
On this panel, you can:
-
Choose another Color Scheme or set all the colors individually (just click each color).
-
Show/Hide Value and Domain Labels and also set the Domain Label Position.
-
Change the chart Orientation.
-
Show/Hide Axes.
-
Change the line width and color for axes and bars.
-
Increase or reduce the separation between bars (up to 50% of the total chart width).
-
Note: Other charts provides different options
-
-
Check both Show Domain Axis and Show Range Axis and apply the graphics again. Now the node chart should look like this:
-
The default domain labels are not very useful, so let's set better labels:
-
On the Node Table (Table Panel), create a new List Column of type String and name it "domain_labels".
-
Edit any of the cells of the created column (double-click it) and type
["Bet. Cent.","Closen. Cent","Clust. Coeff.","Topol. Coeff."]. -
Right-click the same cell and select the option Apply to entire column.
-
Open the chart editor again and select the Options panel.
-
Select "domain_labels" on the Domain Labels Column drop-down button.
-
Select "Up 45^o^" on the Domain Labels Position drop-down button. The labels should look like this now:
-
Several utility functions are available for Discrete Mappings. You can use those functions by right-clicking on any property (shown below.)
-
Mapping Value Generators - Functions in this menu category are value generators for discrete mappings. Users can set values for discrete mappings automatically by these functions.
-
Rainbow and Rainbow OSC - These functions try to assign as diverse a set of colors as possible for each data value.
-
Random Numbers and Random Colors - Randomized numbers and colors.
-
Number Series - Set a series of numbers to the specified mapping. Requires a starting number and increment.
-
Fit label width - This function is only for node width and height. When the node size is unlocked AND Node Width/Height discrete mappings are available, you can fit the size of each node to its label automatically by selecting this function. See the example below:
-
You can set multiple values at once. First, you need to select rows in which you want to change values then right-click and select Edit → Edit Selected Discrete Mapping Values. A dialog pops up and you can enter the new value for the selected rows.
A Property Dependency can be established between different properties. Currently there are two dependencies: Lock node width and height (available in the Node tab) and Edge color to arrows (available in the Edge tab).
-
Lock Node with and height - If this menu item is checked, Node Width and Node Height mappings are ignored and Node Size overrides them. If you want to use the Fit label width function, you need to unlock this.
-
Edge color to arrows - If this menu item is checked then Source Arrow Color and Target Arrow Color for edges are overridden and Edge Color is used in both cases.
There are three kinds of Continuous Mapping Editors. Each of them are associated with a specific properties:
Editor Type Supported Data Type Properties Color Gradient Editor Color node/edge/border/label colors Continuous-Continuous Editor Numbers size/width/opacity Continuous-Discrete Editor All others font/shape/text
Each editor has a common section named Range Setting.
-
Handle Value Box - This box displays the current value for the selected slider handle. You can also directly type the value in this box to move the slider to an exact location.
-
Min/Max Button - Set the overall range of this editor. The first time you open the editor, the Min and Max values are set by the range of the data column you selected, i.e., minimum and maximum value of the column will be set to the range of this editor. You can change this range anytime you want by pressing this button.
-
Add Handle Button - Add a new handle to the editor.
-
Delete Handle Button - Delete the selected handle from the slider widget.
-
Handle Value Editor Button - Edit value assigned to the selected handle.
The Gradient Editor is an editor for creating continuous mappings for colors. To change the color of each region, just double-click the handles (small triangles on the top). A Color gradient will be created only when the editor has two or more handles (see the example below).
1 handle (no gradient) 2 handles
The Continuous-Continuous Editor is for creating mappings between numerical data and numerical properties (size/opacity). To change the value assigned on the Y-axis (the property shown in the example above is node size), drag the red squares or double-click on the squares to directly type an exact value.
The Continuous-Discrete Editor is used to create mappings from numerical column values to discrete properties, such as font, shape, or line style. To edit a value for a specific region, double click on the icon on the track.
All Cytoscape Style settings are initially loaded from a default file
that cannot be altered by users. When users make changes to the
properties, a session_syle.xml file is saved in the session file. This
means that if you save your session, you will not lose your properties.
No other style files are saved during normal operation.
Styles are automatically saved with the session they were created in. Before Cytoscape exits, you will be prompted to make sure you save the session before quitting. It is also possible to save your styles in a file separate from the session file. To do this, navigate to the File → Export → Style... menu option and save the properties as a file. This feature can be used to share styles with other users.
The Cytoscape-native Style format is Style XML. If you want to share Style files with other Cytoscape users, you need to export them to this format.
From version 3.1.0 on, Cytoscape can export Cytoscape.js compatible JSON file. Since Cytoscape.js is an independent JavaScript library, and there are some differences between Cytoscape and Cytoscape.js, not all properties are mapped to JSON. The following properties are not supported by the exporter:
-
Custom Graphics and their locations
-
Edge Bends
-
Nested Networks
-
Network Background (Note: This can be set manually as standard CSS in Cytoscape.js)
The Continuous-Discrete Editor is used to create mappings from numerical data values to discrete properties, such as font, shape, or line style. To edit a value for a specific region, double-click on the icon on the track.
To import existing styles, navigate to the File → Import → Style...
menu option and select a styles.xml (Cytoscape 3 format) file.
Imported properties will supplement existing properties or override
existing properties if the properties have the same name. You can also
specify a style file using the -V command line option. Properties loaded
from the command line will override any default properties. Note that
legacy .props files can only be loaded via the File → Import →
Style... menu option, but not by command line.
Cytoscape's capabilities are not fixed. They can be expanded with apps. They can extend Cytoscape in a variety of ways. One app can have the ability to import data from an online database. Another app could provide a new method for analyzing networks. You can install apps after you have installed Cytoscape. Most apps were made by Cytoscape users like you.
If you're familiar with Cytoscape 2.x, you probably know that Cytoscape apps were called plugins. Starting in Cytoscape 3.0, we are calling them apps. Cytoscape 2.x plugins cannot be used in Cytoscape 3.0.
You can install apps through the App Store or within Cytoscape. In this section, we'll talk about installing apps through Cytoscape. You can learn how to install apps through the App Store here.
To install apps within Cytoscape, go to the menu bar and choose Apps → App Manager. At the top of the App Manager window, make sure you have the Install tab selected.
There are three ways you can find apps:
-
If you know the name of an app you're looking for, enter it in the Search field. The App Manager will list the apps whose names or descriptions match the Search field in the middle panel.
-
If you have a general idea of what sort of app you're looking for, double-click on the apps by tag folder, then click on one of the tags that interests you. The apps with that tag are listed in the middle panel.
-
If you're not sure what sort of app you and want to see everything, click the all apps folder. In the middle pane, you will see a list of all the apps.
When you click on an app in the middle panel, the App Manager shows its short description and icon in the right panel. If you want more information about the app, click the View on App Store button on the bottom-right. If you want to go ahead and install the app, click the Install button.
If you've downloaded an app to your computer, you can install it by clicking the Install from File button on the bottom-left.
You can see a list of all apps you have installed by clicking the Currently Installed tab at the top. When you click on an app in the list, you'll see a description of your app at the bottom. At the bottom, you'll see a couple buttons where you can:
-
Uninstall an app. This deletes the app from your computer. If you want to reinstall the app, you will have to find it again in the Install from App Store tab or in the App Store site and reinstall it there.
-
Disable an app. This temporarily disables the app. The app stays on your computer, but Cytoscape does not load it. You can enable the app by first selecting the disabled app in the list, then click Enable.
The Command Line Tool provides a simple command-line interface to Cytoscape using the Commands API. Any app that registers commands will be available through the Command Tool. The Command Tool provides two main functions: first, a Command Line Dialog is available from Tools → Command Line Dialog, that allows the user to type commands into Cytoscape and see the results in a "Reply Log".
Second, and arguably more useful, it will read script files and execute them. Each line in the script file is a command that is sent to a app. Script files may be entered on the Cytoscape command line using the "-S" flag to Cytoscape, through the File → Run... menu item, or through Tools → Execute Command File.
Cytoscape commands consist of three parts: a command class, or namespace; a command within that namespace; and a series of arguments or options provided as a series of name=value pairs. For example, to import an XGMML format file from the Command Line Dialog or a command script, you would use:
network import file filePath="path-to-file"
where network is the namespace, import file is the command, and there is only one argument: filePath="path-to-file". If there were more arguments they would appear on the same line separated by spaces.
The Command Tool also uses the Command API to provide help. "help" by itself will list all of the command classes (or namespaces) and "help " followed by a namespace will list all of the commands supported by that namespace. Details of a specific command are available by typing "help " followed by the namespace and command (e.g. "help layout force-directed"). The Command Tool registers the "command" namespace and supports a single command: run, which takes a file argument. Here is the help for the command run command from the command namespace:
help command run
command run file=<File>
Similarly, the help for the "network import file" example from above is:
help network import file
network import file RootNetworkList=<ListSingleSelection [Create new network collection]> SourceColumnList=<ListSingleSelection [shared name]> TargetColumnList=<ListSingleSelection [shared name]> defaultInteraction=<String> delimiters=<ListMultipleSelection [,, ;, , \t]> delimitersForDataList=<ListSingleSelection [\|, \, /, ,]> filePath=<File> firstRowAsColumnNames=<boolean> indexColumnSourceInteraction=<int> indexColumnTargetInteraction=<int> indexColumnTypeInteraction=<int> startLoadRow=<int>
Cytoscape allows for merging of both network and table data, through Tools → Merge.
The Advanced Network Merge interface is available from Tools → Merge → Networks... and allows for merging of two or more networks.
-
In the Operation drop-down, select either "union", "intersection" or "difference".
-
Networks available for merge are listed under Networks to merge. Select a network from the list and click the right arrow to transfer the network to Selected networks. Click Merge to continue. The merged network will be displayed as a separate network.
The Advanced Network Merge interface includes an expandable Advanced Network Merge panel, where you can specify the details of how to merge the networks. The options available here are:
-
Matching columns: This specifies the network columns that should be used for merging. Typically, the "name" column or some other column containing identifier information is used here.
-
How to merge columns?: A table lets the user specify for each of the individual network columns, what the corresponding column in the merged network should be named and its data type.
NetworkAnalyzer computes a comprehensive set of topological parameters for undirected and directed networks, including:
-
Number of nodes, edges and connected components.
-
Network diameter, radius and clustering coefficient, as well as the characteristic path length.
-
Charts for topological coefficients, betweenness, and closeness.
-
Distributions of degrees, neighborhood connectiveness, average clustering coefficients, shortest path lengths, number of shared neighbors and stress centrality.
NetworkAnalyzer also constructs the intersection, union and difference of two networks. It supports the extraction of connected components as separate networks and the removal of self-loops.
To run NetworkAnalyzer, select Tools → NetworkAnalyzer → Network Analysis → Analyze Network.
The NetworkAnalyzer will determine whether your network contains directed or undirected edges. At this point, you can choose to ignore edge direction information.
When results are calculated, they will appear in the Results Panel.
The results have multiple tabs. Details on the network parameters can be found here.
-
Simple Parameters
-
Node Degree Distribution
-
Avg. Clustering Coefficient Distribution
-
Topological Coefficients
-
Shortest Path Distribution
-
Shared Neighbors Distribution
-
Neighborhood Connectivity Distribution
-
Betweenness Centrality
-
Closeness Centrality
-
Stress Centrality Distribution
You can also save the statistics for later use by using the Save Statistics button.
An exhaustive topological analysis of very large networks can be a time consuming task. The computation of local parameters for the nodes is significantly faster than the computation of global (path-related) parameters. Examples of local parameters are node degree, neighborhood connectivity, topological and clustering coefficients. Betweenness and closeness centralities, as well as stress, are global parameters.
NetworkAnalyzer provides the Analyze Subset of Nodes option for computing local parameters for a subset of nodes. If one or more nodes in the network are selected before starting an analysis, only the sub-network induced by the selected nodes is analyzed. Moreover, only local parameters are computed. Shared neighbors distribution and shortest path lengths distribution, among others, are not displayed in the results.
The Batch Analysis option is used to perform topological analysis on all networks stored in specific directory, using all possible interpretations for every network. Batch analysis consists of three simple steps:
-
Selecting directories: The user selects the input and output directories. The input directory should contain network files that can be loaded into Cytoscape. Sub-directories of the input directory are not considered. The output directory is the one that will contain all analysis results after the batch analysis. In order to avoid file overwriting, NetworkAnalyzer requires that the output directory is empty (contains no files) before the batch analysis starts.
-
Analysis: NetworkAnalyzer scans the input directory and loads all supported networks into Cytoscape, one at a time. Each loaded network is inspected and then it is analyzed considering all possible interpretations for it. The analysis step is complete after all networks are analyzed. Note that depending on the number of networks and their sizes, this might be a very time-consuming step.
-
Inspection of results: After the analysis is complete, the button Show Results is enabled, and it displays the results dialog. The dialog contains a table of all topological analyses performed. Every row in the results table lists the loaded network, its interpretation and the resulting network statistics file that was saved in the output directory. By clicking on a network name and on statistics file name, the user can load the network and topology analysis results, respectively.
Existing network statistics can be loaded from a file saved previously in NetworkAnalyzer.
The Plot Parameters dialog offers a possibility to plot two parameters against each other. The parameters to be plotted can be chosen from two drop-down menus. The Table Column 1 menu provides the values for the domain/category axis, and the Table Column 2 menu specifies the values for the range/value axis. The plot is updated each time a different parameter is selected in one of the menus.
NetworkAnalyzer computed parameters can be visualized as node/edge size and color, if the Store node / edge parameters in node / edge table option in NetworkAnalyzer Settings is enabled. Parameters loaded from a .netstats file cannot be visualized because the network itself is not stored in the network statistics file. If, after performing topological analysis, the network is modified by introducing or removing nodes or edges, it is recommended (and sometimes required) to run NetworkAnalyzer again before visualizing any parameters.
The visualization is initiated by the Generate Style from Statistics... menu option. There are two ways of mapping computed parameters.
-
Map to node / edge size: The computed parameter is mapped to the size of the nodes or edges. Mapping can be straight or inverse, that is, low parameter values can be mapped to small sizes or to large sizes. The smallest node size is set to 10 and the largest one to 100. Regarding edges, size reflects the edge line width and varies between 1 and 8. Refer to the Styles section for details.
-
Map to node / edge color: A computed parameter is mapped to the color of the nodes or edges. Two mapping styles are possible - mapping low parameter values to bright colors or to dark colors. By default, the brightest color is green and the darkest color is red. The mapping also uses a middle (intermediate) color, which allows for fine-grained perception of differing values through the color gradient. The default middle color is yellow.
The following settings can be configured by the user:
-
Store node / edge parameters in node / edge table: For every node in a network, NetworkAnalyzer computes its degree (in- and out-degrees for directed networks), its clustering coefficient, the number of self-loops, and a variety of other parameters. NetworkAnalyzer also computes edge betweenness for each edge in the network. If the respective options are enabled, NetworkAnalyzer can stores the computed values as columns of the corresponding nodes and edges. This enables the users to use the values in Styles or to select nodes or edges based on the values. A complete list of the computed node and edge columns is available here.
-
Use expandable dialog interface for the display of network statistics: If this option is enabled, analysis results are presented in a window in which all charts are placed below each other in expandable boxes. If this option is disabled, analysis results are presented in a window that contains tabs for the group of simple parameters and for every complex parameter (default). Users who wish to simultaneously view two or more complex parameters of one network, should enable this option.
-
NetworkAnalyzer allows the user to change the default colors of parameter visualization.
-
Background color for parameter visualization: The color of the background in the network view. It is initially set to the default Cytoscape background color.
-
Bright color to map parameters: This color defines the brightest color that parameters can be mapped to. By default its value is green.
-
Middle color: This color defines the intermediate color, that parameters can be mapped to. By default its value is yellow.
-
Dark color: This color defines the darkest color that parameters can be mapped to. By default its value is red.
-
-
Location of the help documents: URL of the original help web page for NetworkAnalyzer. This also enables the local download and storage of this help page.
NetworkAnalyzer allows for the creation of sub-networks of connected components. The user selects a number of connected components from a list and each selected components is visualized as a sub-network.
The Cytoscape properties editor, accessed via Edit → Preferences →
Properties…, is used to specify default properties. Any changes made
to these properties will be saved in .props files under the
CytoscapeConfiguration subdirectory of the user's home directory.
Cytoscape properties are configurable using the Add, Modify and Delete buttons as seen below.
App properties may also be edited in the same way as editing Cytoscape
properties. For example, to edit the properties of Linkout, select
'linkout' from the combobox of the Preferences Editor. Some apps may
store properties inside session files in addition to (or instead of)
storing them in the CytoscapeConfiguration directory.
Cytoscape contains a pre-defined list of bookmarks, which point to sample network files located on the Cytoscape web server. Users may add, modify, and delete bookmarks through the Bookmark manager, accessed by going to Edit → Preferences → Bookmarks….
There are currently several types of bookmarks (based on data categories), including network and table. Network bookmarks are URLs pointing to Cytoscape network files. These are normal networks that can be loaded into Cytoscape. Table bookmarks are URLs pointing to data table files.
You can define and configure a proxy server for Cytoscape by going to Edit → Preferences → Proxy Settings….
After the proxy server is set, all network traffic related to loading a
network from URL will pass through the proxy server. Cytoscape apps use
this capability as well. The proxy settings are saved in
cytoscape3.props. Each time you click the OK button after making a
change to the proxy settings, an attempt is made to connect to a well
known site on the Internet (e.g., google.com) using your settings. For
both success and failure you are notified and for failure you are given
an opportunity to change your proxy settings.
If you no longer need to use a proxy to connect to the Internet, simply set the Proxy type to "direct" and click the OK button.
The configuration of Cytoscape group view may also be edited through Edit → Preferences → Group Preferences….
Linkout provides a mechanism to link nodes and edges to external web resources within Cytoscape. Right-clicking on a node or edge in Cytoscape opens a popup menu with a list of web links.
The external links are specified in a linkout.props file which is
internal to Cytoscape. The defaults include a number of links such as
Entrez, SGD, iHOP, and Google, as well as a number of species-specific
links. In addition to the default links, users can customize the
External Links menu and add (or remove) links by editing the linkout
properties (found under Edit → Preferences → Properties...).
External links are listed as ‘key’-‘value’ pairs in the
linkout.props file where key specifies the name of the link and
value is the search URL. The LinkOut menus are organized in a
hierarchical structure that is specified in the key. Linkout key terms
specific for nodes start with the keyword nodelinkouturl, for edges
this is edgelinkouturl.
For example, the following entry:
nodelinkouturl.Model Organism DB.SGD (yeast)=http://www.yeastgenome.org/cgi-bin/locus.fpl?locus=%ID%
places the SGD link under the Model Organism DB submenu. This link will appear in Cytoscape as:
In a similar fashion one can add new submenus.
The %ID% string in the URL is a place-holder for the node label.
When the popup menu is generated this marker is substituted with the
node label. In the above example, the generated SGD link for the YNL050C
protein is:
http://www.yeastgenome.org/cgi-bin/locus.fpl?locus=YNL050C
If you want to query based on a different column, you need to specify a different node label using Styles.
For edges the mechanism is much the same; however here the placeholders
%ID1% and %ID2% reflect the source and target node label
respectively.
Currently there is no mechanism to check whether the constructed URL query is correct and if the node label is meaningful. Similarly, there is no ID mapping between various identifiers. For example, a link to NCBI Entrez from a network that uses Ensembl gene identifiers as node labels will produce a link to Entrez using the Ensembl ID, which results in an incorrect link. It is the user's responsibility to ensure that the node label that is used as the search term in the URL link will result in a meaningful link.
The default links are defined in a linkout.props file contained inside
the Linkout JAR bundle under the
framework/system/org/cytoscape/linkout-impl subdirectory of the
Cytoscape installation. These links are normal Java properties and can
be edited by going to Edit → Preferences → Properties... and
selecting linkout from the box (shown below). Linkouts can be modified,
added or removed using this dialog; however, note that the modifications
would not be stored in the file. To change a URL permanently, you would
need to edit the linkout.props file directly.
In addition, new links can be defined when starting Cytoscape from
command line by specifying individual properties. The formatting of the
command is cytoscape.sh -P [context_menu_definition]=[link].
context_menu_definition specifies the context menu for showing the
linkout menu item. The structure of this definition is "." separated and
the first item needs to be either nodelinkouturl or edgelinkouturl.
The former will add the linkout item as a node context menu and the
latter will add it as an edge context menu. The rest of the definition
would define the hierarchy of the menu.
For instance this command:
cytoscape.sh -P nodelinkouturl.yeast.SGD=http://db.yeastgenome.org/cgi-bin/locus.pl?locus\=%ID%
will add this menu item:
To remove a link from the menu, simply delete the property using Edit → Preferences → Properties... and selecting commandline. Linkouts added in the command line will be available for the running instance of Cytoscape.
Panels are floatable/dockable panels designed to cut down on the
number of pop-up windows within Cytoscape and to create a more unified
user experience. They are named based on their functions -- Control
Panel, Table Panel, Tool Panel and Result Panel. The
following screenshot shows the file galFiltered.sif loaded into
Cytoscape, with a force-directed layout, Align and Distribute tools
showing in the Tool Panel, and with results from Network Analyzer
(Tools → Network Analysis → Analyze Network). The Control Panel
(at the left-hand side of the screen) contains the Network Manager,
Network Overview, Styles and Select tabs. On the bottom of the panel,
there is another panel called Tool Panel. In the Table Panel,
the Node Table is shown. In addition, analysis results from Network
Analyzer are shown in Result Panel (at the right-hand side).
The user can then choose to resize, hide or float Panels. For example, in the screenshot below, the Network, Table and Results panels are floating and the Tool Panel is hidden:
Cytoscape includes four Panels: the Control Panel on the left, Tool Panel on the bottom of the Control Panel, the Table Panel on the bottom right, and the Result Panel on the right. By default, only the Control Panel and the Data Panel will appear. The Result Panel may appear, depending on the mix of Cytoscape apps that you currently have installed. The Tool Panel will appear when you select the following commands under the Layout menu: Rotate, Scale, and Align and Distribute.
All panels can be shown or hidden using the View → Show/Hide functions.
In addition, Panels can be floated or docked using icon buttons at the
top right corner of each Panel. The Float Window control
Cytoscape 3.0 retains the rendering engine found in version 2.8. It is to be able to display large networks (>10,000 nodes), yet retain interactive speed. To accomplish this goal, a technique involving level of detail (LOD) is being used. Based on the number of objects (nodes and edges) being rendered, an appropriate level of detail is chosen. For example, by default, node labels (if present) are only rendered when less than 200 nodes are visible because drawing text is a relatively expensive operation. This can create some unusual behavior. If the screen currently contains 198 nodes, node labels will be displayed. If you pan across the network, such that now 201 nodes are displayed, the node labels will disappear. As another example, if the sum of rendered edges and rendered nodes is greater than or equal to a default value of 4000, a very coarse level of detail is chosen, where edges are always straight lines, nodes are always rectangles, and no anti-aliasing is done. The default values used to determine these thresholds can be changed by setting properties under Edit → Preferences → Properties....
Low LOD vs High LOD
Network with Low LOD Network with High LOD
With low LOD values, all nodes are displayed as square and anti-aliasing is turned off. With high LOD values, anti-aliasing is turned on and nodes are displayed as actual shape user specified in the Style.
NOTE: The greater these thresholds become, the slower performance will become. If you work with small networks (a few hundred nodes), this shouldn't be a problem, but for large networks it will produce noticeable slowing. The various thresholds are described below.
render.coarseDetailThreshold If the sum of rendered nodes and rendered edges equals to or exceeds this number, a very coarse level of detail will be chosen and all other detail parameters will be ignored. If the total number of nodes and edges is below this threshold, anti-alias will be turned on; this value defaults to 4000. render.nodeBorderThreshold If the number of rendered nodes equals to or exceeds this number, node borders will not be rendered; this value defaults to 400. render.nodeLabelThreshold If the number of rendered nodes equals to or exceeds this number, node labels will not be rendered; this value defaults to 200. render.edgeArrowThreshold If the number of rendered edges equals to or exceeds this number, edge arrows will not be rendered; this value defaults to 600. render.edgeLabelThreshold If the number of rendered edges equals to or exceeds this number, edge labels will not be rendered; this value defaults to 200.
When printing networks or exporting to formats such as PostScript, the highest level of detail is always chosen, regardless of what is currently being displayed on the screen.
If you want to display every detail of the network regardless of LOD values, you can toggle to full details mode by View → Show Graphics Details (or CTR+SHIFT+F on Windows/Linux, Command+SHIFT+F for Mac). This option forces the display of all graphics details. If the network is large, this option slows down rendering speed. To hide details, select the menu item again (View → Hide Graphics Details).
When you finish your data analysis and visualization, you need to publish your data to share the results. Cytoscape has several options to do it, with most options suitable for Cytoscape users and other options suitable for programmers wanting to create unusual or complex network viewers.
-
A session file
-
A static image
-
An interactive web application
-
Full web application
-
Simple network view (for web application developers)
The easiest way to share your results with others is simply save everything as a session file (which is a zipped session archive). You can save your current session by clicking Save Session icon. You can save to a thumb drive, a shared file system, or even a cloud drive directory such as Dropbox -- if you save to a shared drive, beware not to have two people the session file with Cytoscape at the same time, as unpredictable results may occur.
Cytoscape can generate publication-quality images from network views. By selecting File → Export → Network View as Graphics..., you can export the current network view into the following files:
-
JPG
-
PNG
-
PS (Post Script)
-
SVG
-
PDF
We recommend using PDF for publications because it is a standard vector graphics format, and it is easy to edit in other applications such as Adobe Illustrator.
For PDF export, there is an option to Export Texts as Fonts. This option does not work for two-byte characters such as Chinese or Japanese. To avoid corrupted texts in the exported images, please uncheck this option when you publish networks including those non-English characters.
The Web is an excellent platform for data sharing and collaboration, and Cytoscape provides a number of ways to publish your network on the web, with each choice representing tradeoffs between ease, simplicity, and customization options. All solutions leverage the cytoscape.js drawing library, and so enable not only viewing but also Cytoscape-style interactive browsing of networks and attributes.
The simplest choice is CyNetShare, where you save your network (and optionally a style) on a public file system, then interactively view the network in a browser. Like Google Maps, you can generate and publish a URL that allows collaborators to also view your network.
Alternatively, Cytoscape can generate an entire web site showing a single page containing the viewer with your network pre-loaded. You can load this directly onto your own web server to become part of your web site.
Finally, Cytoscape can generate the skeleton of a web site for further customization by JavaScript programmers.
These features are available as Export menu items under the File menu, and are described in sections below.
For example, here is a network in Cytoscape:
Here is the same network as an interactive web visualization:
Note that web browsers can render small networks (e.g., 1000 nodes) quickly and effectively, but attempting to render large ones (e.g., 5000 nodes) will take a very long time.
CyNetShare is a browser-based web application that renders JSON-formatted networks and attributes saved in public directories. Optionally, you can specify visual styles that the web application will use to draw your network as it appears in Cytoscape. CyNetShare is similar to Google Maps in that once you have loaded your network and have tweaked its appearance to suit, you can have CyNetShare generate a new URL that you can e-mail or post as a link on your own web site. That URL will bring up CyNetShare preloaded with your network for anyone to see.
To use CyNetShare:
-
Select File → Export → Network and View... to export your network to a public directory. Choose the Cytoscape.js JSON (*.cyjs) export file format.
-
Optionally, select File → Export → Style... to export your style settings. Choose the Style for cytoscape.js (*.json) export file format.
-
Use your public directory system to determine direct URLs for the files you exported.
-
Start CyNetShare
-
Enter the network's URL as the Graph URL.
-
Optionally, enter the style's URL.
-
Click the Visualize button.
The CyNetShare User Guide is provided on the CyNetShare web page:
Note that if you specify a style URL, the style is added to the list of styles available in CyNetShare's Visual Style dropdown, and you can apply the style by selecting it in the list.
Note that the mechanics of generating a URL for a file in a public directory system is a fast moving topic. Until recently, systems like Dropbox (and others) allowed users to create a URL that resolved directly to a file -- a "direct" URL would be appropriate for use with CyNetShare. As of this writing, some public directory systems (e.g., Dropbox) generate "shareable" URLs instead, which resolve to a web page that allows file download -- a "shareable" URL makes CyNetShare hang. Systems that offer "shareable" URLs may offer "direct" URLs as part of their premium (or Pro) package. To tell if your public directory system generates a "direct" URL, have it generate a URL for a file, then paste the URL into the address field of a browser and observe whether the browser displays the file itself (good!) or a download page for the file (bad!).
A simple strategy for always getting a "direct" URL is to store your file in a public directory served up by a web server, if you have access to one -- a URL served by a web server might appear as: .
Alternately, you can use Gist to create a shareable document having a "direct" URL. To try this:
-
Use Cytoscape to generate your network as a .cyjs file on your local disk
-
Use an editor to open the file and copy its contents to the clipboard
-
Use a web browser to surf to Gist
-
Paste the contents into a Gist document
-
Select Create public Gist
-
Select Raw to place the "direct" URL into your browser's address field
-
Use that URL with CyNetShare
The full page export option is designed for users who want to publish their network as a complete single-page application. Cytoscape creates a zip archive containing a complete JavaScript-based web application that works as a basic viewer (like CyNetShare) for Cytoscape-generated network visualizations. You can unzip the archive onto a web server (or your PC) and view the network via a web browser on PCs and tablets.
To generate an entire web page as a zip archive, select File → Export → Network View(s) as Web Page ....
To view the web page, unzip the archive into a folder on your PC or web server. The folder will contain an index.html file, the network data, and other files. You can open the index.html file in your browser (usually from your browser's File → Open menu item.)
Depending on your browser's security settings, you may not be able to open the index.html file directly if it is stored on your PC -- you may need to start a web server on your PC. An easy way to set up a local web server is by running the Python simple HTTP server. If you have Python installed on your machine, just go into the web archive folder and type:
python -m SimpleHTTPServer 8000
and use your browser to open:
http://localhost:8000/
Testing the archive on your PC will serve as an easy test of the web page, but to publish it to collaborators, you should unzip your archive onto a web server.
Here is an example exported file from Cytoscape:
Note that because Cytoscape uses the latest HTML5-based web technologies, it cannot support older or non-conformant web browsers such as Internet Explorer. We strongly recommend using the latest version of modern web browsers such as Google Chrome, Mozilla Firefox, or Apple Safari.
The Simple Network View export option is designed for users who want to publish their data as a complete single-page application, but are interested in customizing the web viewer application themselves. Cytoscape creates a zip archive containing a partial JavaScript-based web application based on the cytoscape.js library and including simple "boilerplate" code and the current network view. The user can create a custom viewer by customizing this code.
To generate an entire web page as a zip archive, select File → Export → Network View(s) as Web Page ..., and choose the Simple viewer for current network only format.
For instructions on testing the customized web application, see Generating a Full Web Application above.
The code generated by Cytoscape for the Full Web Application and the Simple Network View web applications is minimalistic. While you can directly modify this code yourself to create your own page design or add new features, the modifications will apply to a single exported network. If you are a web application developer, you can change the application code generated for all exports by editing HTML5 template code resource files in your ~/CytoscapeConfiguration/web directory:
In this folder, you can find full and simple sub directories corresponding to Full Web Application and the Simple Network View described above.
To build these project, you need the following tools installed:
-
Node.js
-
gulp
-
grunt
This is an AngularJS based web application built with grunt (at least for now -- we have plans to migrate to gulp). Source code and more documentations are available here:
To build the project into dist directory, type:
grunt
This template is generated by a simple gulp project. The source code is available here:
To build the project into dist directory, type:
gulp
Once you have your own builds, you can deploy your templates by replacing the contents of full and simple with your own builds.
Cytoscape.js is a JavaScript library for interactive network visualization. It is a building block for web applications and is NOT a complete web application. If you have network data sets and want to share visualizations created with Cytoscape, you can build your own website using Cytoscape.js and this new Export to Cytoscape.js feature.
-
A network visualized with Cytoscape 3.1.0
-
Same network exported to Cytoscape.js (Rendered with Google Chrome and Cytoscape.js 2.0.3)
-
Interactive example (galFiltered.sif rendered with Cytoscape.js 2.0.3)
Although Cytoscape and Cytoscape.js are two completely independent software packages, they are sharing higher level concepts. The following is the list of similarities and differences between the two:
-
Desktop application for network visualization written in Java programming language
-
Needs desktop or laptop computers to run
-
Users have to install Java runtime
-
High performance application for large scale network analysis and visualization
-
Expandable by Apps
-
Use Styles to map data to network properties, such as node color, edge width, node shape, etc.
-
A JavaScript library for network visualization, NOT a complete web application nor mobile app
-
Runs on most of modern web browsers, including tablets and smart phones
-
No plugins are required to run. Modern web browser is the only requirement
-
Need to write code to set up your web site or web application
-
Expandable by Extensions
-
Use CSS-based Styles to map data to network properties
In a long term, Cytoscape and Cytoscape.js will be more integrated, and as the first step Cytoscape now supports reading and writing Cytoscape.js network/table JSON files. In addition, Cytoscape can convert Styles to Cytoscape.js Style object.
Since Cytoscape.js is a JavaScript library, its basic data exchange format is JSON (JavaScript Object Notation). All of these import/export functions are part of standard Cytoscape user interface, and you can read/write Cytoscape.js JSON files just like other standard files such as SIF.
Cytoscape.js stores network data (graph) and its data table in the same object. Cytoscape writes such data complex as JSON, i.e., both network and data tables will be exported as a single JSON file. You can select a network and export it from File → Export → Network.
Cytoscape only supports one of the Cytoscape.js supported JSON formats, which is:
{
elements:{
nodes:[],
edges:[]
}
}
SUID will be used as unique identifier for nodes and edges in the JSON. For more information about this format, please visit Cytoscape.js web site.
Cytoscape creates JSON file directly from data table and tries to extract as much data as possible from the original table. However, since table column names will be directly converted into JavaScript variable names, invalid characters will be replaced by underscore (_):
-
Original Data Table Column Names:
Gene Name KEGG.pathway -
The Names above will be replaced to:
Gene_Name KEGG_pathway
You should be careful when you plan to use this feature for data roundtrip: from Cytoscape to Cytoscape.js back to Cytoscape. For such use cases, we strongly recommend to use JavaScript-safe characters only in your table column names. Naming your columns only with alphanumeric characters and underscore (_) is the best practice. (For actual data entries, all characters are allowed. This restriction is only for table column names.)
Cytoscape and Cytoscape.js are sharing a concept called Style. This is a collection of mappings from data point to network property. Cytoscape can export its Styles into CSS-based Cytoscape.js JSON.
You can export all Styles into one JSON file from File → Export → Style and select Cytoscape.js JSON as its format.
Cytoscape.js does not support all of Cytoscape Network Properties. Those properties will be ignored or simplified when you export to JSON Style file.
Currently, the following Visual Properties will not be exported to Cytoscape.js JSON:
-
Custom Graphics and its positions
-
Edge Bends
-
Tooltips
-
Node Label Width
-
Node Border Line Type
-
Arrow Colors (they are always same as edge color)
The following two charts detail the compatibility of specific Cytoscape features in Cytoscape.js.
Cytoscape.js network JSON files can be loaded from standard Cytoscape file menu: File → Import → Network →. Both File and URL are supported.
Although Cytoscape can export networks, tables, and Style as Cytoscape.js-compatible JSON, users have to write some JavaScript code to visualize the data files with Cytoscape.js. Details of web application development with Cytoscape.js is beyond the scope of this document. If you need examples and tutorials about web application development with Cytoscape.js, please visit the following web site:
If you have questions and comments about web application development with Cytoscape and Cytoscape.js, please send yours to our mailing list.
Due to limitations of JVM languages and OSGi, this feature is deprecated. Please use cyREST instead.
Scripting language support is an experimental feature in Cytoscape 3. Java runtime environment comes with built-in scripting language support and Cytoscape uses it to execute tasks written in standard languages such as JavaScript.
This feature is for advanced users. If you just want to automate some basic tasks, such as loading network, you can use Command instead.
To use Scripting feature, you need to start Cytoscape from console. In Cytoscape application directory, you can find cytoscape.sh (for Windows, cytoscape.bat). This is a shell script to start Cytoscape from command line.
Once you run Cytoscape from command line, you can use Apache GOGO OSGi Shell. If you type:
help cytoscape:script
the shell displays help message how to use script files in Cytoscape.
JavaScript is supported by default. To run your script, type:
cytoscape:script javascript SCRIPT_FILE_NAME
also, you can pass optional command line arguments.
// Generate a complete graph
// Extract command-line arguments
var numberOfNodes = args[0];
if(numberOfNodes == null)
numberOfNodes = 10;
// Create new network
var newNetwork = cyAppAdapter.getCyNetworkFactory().createNetwork();
newNetwork.getRow(newNetwork).set("name", "Complete Graph with " + numberOfNodes + " Nodes");
// Register the new network to Network Manager
cyAppAdapter.getCyNetworkManager().addNetwork(newNetwork);
// Add nodes
var nodes = [];
for( i = 0; i < numberOfNodes; i++) {
var nodeName = "Node " + i;
var node = newNetwork.addNode();
newNetwork.getRow(node).set("name", nodeName);
nodes[i] = node;
}
// Add edges
var edgeCount = 0;
for( i = 0; i < numberOfNodes; i++) {
var source = nodes[i];
for( j = 0; j < numberOfNodes; j++) {
var target = nodes[j];
if(newNetwork.containsEdge(source, target) == false &&
newNetwork.containsEdge(target, source) == false && j != i) {
var edge = newNetwork.addEdge(source, target, true);
newNetwork.getRow(edge).set("name", "Edge " + edgeCount++);
newNetwork.getRow(edge).set("interaction", "interacts_with");
}
}
}
This script is available here:
By default, cyAppAdapter object will be injected, just like Simple Apps. You can access basic functions from this object. For more information, please read Cytoscape API document.
Warning: This method directly adds ALL classes provided by the library to your classpath. Maybe there are some side-effects to other OSGi Apps.
In theory, you can use full features of scripting languages written on top of JVM by adding JSR 223 (Scripting for the Java Platform) compatible Scripting Engines to Cytoscape runtime. The description below is how to add new scripting languages to Cytoscape 3.
Scripting feature in Cytoscape is very simple. Cytoscape just searches available JSR 223 compatible Scripting Engines and use specified Engine when user runs a script file. To add a new Scripting Engine, simply add the JAR files to CYTOSCAPE_APPLICATION_DIR/framework/lib/ext/. For example, if you want to add Python (Jython) scripting engine, copy the standalone jar file to this directory:
Now it's ready to use. Just specify scripting engine in the command line arguments.
Support for Python can be added by importing Jython Scripting Engine. Simply copy standalone version of JAR file to the ext directory. Access to cyAppAdapter is also available for Python.
This script creates a complete graph with 20 nodes:
#
# Python Script to generate complete graph
#
NUM_NODES = 20
print "Creating complete graph with " + str(NUM_NODES) + " nodes..."
new_network = cyAppAdapter.getCyNetworkFactory().createNetwork()
new_network.getRow(new_network).set("name", "Complete Graph Created by Python Script")
cyAppAdapter.getCyNetworkManager().addNetwork(new_network);
# Add nodes
nodes = [];
for i in range (NUM_NODES):
node_name = "Node " + str(i)
node = new_network.addNode()
new_network.getRow(node).set("name", node_name)
nodes.append(node)
# Add edges
edge_count = 0;
for source in nodes:
for target in nodes:
if new_network.containsEdge(source, target) is False \
and new_network.containsEdge(target, source) is False \
and source is not target:
edge = new_network.addEdge(source, target, True)
edge_count = edge_count + 1
new_network.getRow(edge).set("name", "Edge " + str(edge_count))
new_network.getRow(edge).set("interaction", "interacts_with")
To run, type:
cytoscape:script python /ABS_PATH_TO_SCRIPT/complete_graph.py
The original Clojure distribution does not contain JSR 223 implementation. You need to build this library to run Clojure scripts.
; Node Names - sequence of integers.
(def nodenames (take 500 (range)))
; Create new network and register it to manager
(def newnet(. (. cyAppAdapter getCyNetworkFactory) createNetwork))
(.. newnet (getRow newnet) (set "name" "Random Network Created by Clojure Script"))
(.. cyAppAdapter (getCyNetworkManager) (addNetwork newnet))
; Create nodes
(doseq [nodename nodenames]
(.. newnet (getRow (. newnet addNode))
(set "name" (str nodename))))
; Add random edges
(def nodes (. newnet getNodeList))
(doseq [node nodes]
(. newnet addEdge node (rand-nth nodes) false))
To display log messages, use this command:
tail -f ~/CytoscapeConfiguration/3/framework-cytoscape.log
Since dependencies are added in Classpath, you cannot access static methods in Cytoscape core API. The only way to access those is put actual API JAR files to system classpath.
We respect the privacy of all Cytoscape users, and we do not collect any information on Cytoscape users except in the situations listed below. In no case do we attempt to tie any of this information back to a user, nor do we give, share, sell, or transfer this information to any third party unless required by law. We use this information only in the aggregate to generate statistics to assist in securing continued funding for Cytoscape.
-
On the Cytoscape download web page, we request your name, organization, and e-mail address, though you are not obligated to provide them. When you actually download Cytoscape, we log the date and time and the IP address to which we deliver Cytoscape.
-
For a news feed fetched for display on the Cytoscape Welcome screen, we log the date and time the news was fetched, and the IP address and browser signature for the workstation running Cytoscape.
This policy may change from time to time, and if it does, we will notify you via the Cytoscape Welcome screen news feed and via our normal mass notification media. We will also update this section of the user manual.
Note that some internal Cytoscape Apps and Apps available through the Cytoscape App Store connect with third party services via the Internet. Once an App links to such a service, you are subject to the privacy policy of that service.