#Title
Authors Affiliations
Suggested figures limit: 2 figures (please follow figure preparation guidelines from Nature Biotech, (http://www.nature.com/nbt/authors/submit/index.html#figures).
##Summary Sentence (suggested limit 30 words)
Include 1 short sentence description of your method that can be used in a main text table.
##Background/Introduction (suggested limit 200 words)
Please try to address the following points:
- What is the motivation for your approach? This will include any previous work and observations that you have made about the NIEHS-NCATS-UNC DREAM data to suggest your approach is a good one. Provide the reader with an intuition of how you approached the problem
- What is the underlying methodology used (e.g., SVM or regression)?
- Where there any novel approaches taken in regards to feature selection, data imputation, ranking, etc?
##Methods (suggested limit 800 words)
The methods should cover a full description of your methods so a reader can reproduce them. Please cover how you processed the data, if any data was imputed or manipulated in any way (e.g., you mapped data onto pathways or combined different datasets), the underlying algorithm, any modifications to the underlying method of importance, the incorporation of outside data, and the approach to predict submitted data. If you submitted multiple predictions, please specify which is the difference among them (e.g. only parameters tuning or different algorithms). If needed, you can decide to write one sub-paragraph for each submission.
##Conclusion/Discussion (suggested limit 200 words)
This section should include a short summary and any insights gained during the algorithm. For example, which dataset was most informative? You can include future directions. You may also add some discussion on the general performance of your methodology (if you wish) and if there were pitfalls, what are they?
##References (suggested limit 10 references)
##Authors Statement Please list all author’s contributions
##Synapse Instructions
Markdown source code for this page is available in synXXXXXXX or as a Gist . Help is also available from the Formatting Guide in the Wiki editor.
Synapse Wikis may contain figures, tables and equations. We show examples below.
###Images There are two ways to include images. The first is a light-weight way appropriate for logos: ${image?fileName=sage%5Flogo%2Egif&align=None&scale=100}
For figures derived from data, uploading them as a Synapse entity enables features like versioning and provenance tracking: ${image?synapseId=syn2318284&align=None&scale=67} Example figure. See files source_code/project_setup_example.R and source_code/ec10_distributions.R
###Source code Source code can be embedded in the page. For example, the code that produced the figure above is:
library(synapseClient)
library(reshape)
library(ggplot2)
synapseLogin()
training_entity <- synGet('syn1917742')
training <- read.table(getFileLocation(training_entity), header=T, row.names=1)
sds <- apply(training, 2, sd)
df <- melt(training[,sds>0.2818], variable_name='compound')
colnames(df)[2] <- 'EC10'
png(file='ec10_distributions.png', height=400, width=700, units='px')
p <- ggplot(df, aes(x=EC10, fill=compound))
+ geom_density(alpha=.3)
+ ggtitle('The distribution of EC10 values for the ten highest variance compounds')
p###Tables Tables are defined in markdown by using the pipe character:
| Submissions | RMSE | Rank |
|---|---|---|
| synXYYYZZZ | 0.03382 | 1 |
| syn2318279 | 0.10951 | 7 |
###Equations Equations can be entered in LaTeX format: $$ \begin{aligned} \hat{\beta} = \underset{\beta}{argmin}\left ( \left | \mathbf{y}-\mathbf{X}\beta \right |^{2}+\lambda_{2}\left | \beta \right |^{2} + \lambda_{1}\left | \beta \right |_{1} \right ) \end{aligned} $$