gregcaporaso

Some exploration into how the pandas DataFrame could be a useful underlying representation for QIIME's metadata mapping files.

gregcaporaso

#!/usr/bin/env python
# File created on 26 Feb 2014
from __future__ import division

__author__ = "Greg Caporaso"
__copyright__ = "Copyright 2014, The QIIME Project"
__credits__ = ["Greg Caporaso"]
__license__ = "GPL"
__version__ = "1.8.0-dev"
__maintainer__ = "Greg Caporaso"

gregcaporaso

#!/usr/bin/env python
# File created on 01 Dec 2011
from __future__ import division

__author__ = "Greg Caporaso"
__copyright__ = "Copyright 2011, The QIIME project"
__credits__ = ["Greg Caporaso"]
__license__ = "GPL"
__version__ = "1.3.0-dev"
__maintainer__ = "Greg Caporaso"

gregcaporaso

#!/usr/bin/env python

from sys import argv
from random import random
from cogent.parse.fastq import MinimalFastqParser
from cogent.draw.distribution_plots import generate_box_plots
from qiime.quality import ascii_to_phred33
from qiime.util import qiime_open

def fastq_quality_plots(fastq_records,

gregcaporaso

#!/usr/bin/env python
# Authors: Greg Caporaso, John Chase
# Questions: Contact [email protected]

# Step 1: Generate lists of pre/post sample ids on a per-individual basis
# qiime.group.extract_per_individual_states_from_sample_metadata 
# will let you generate a dict of individual id to (pre sample-id, post sample-id)

# Step 2: Extract distances for pre/post sample ids
# qiime.parse.parse_distmat_to_dict 

gregcaporaso

These files were used while developing pyqi's Getting Started tutorials. See those documents for usage examples.

gregcaporaso

Taxonomic specificity of sequences in Greengenes

Here I'm creating a hash of expected 515F/806R amplicons from the Greengenes OTUs (for a couple of different sizes of OTUs), and comparing the uniqueness of sequences with the number of different taxonomic identities at each level.

There are basically three categories of sequences:

1. those that are unique, and therefore can only map to a single taxa
2. those that are not unique, but still only map to a single taxa
3. those that are not unique, and map to multiple taxa.

gregcaporaso

                                                                                                                                             41,9          All
#!/usr/bin/env python
# Author: Greg Caporaso

from os.path import join, isdir
from glob import glob

base_in_dir = "/home/caporaso/analysis/short-read-tax-assignment/data/qiime-mock-community/multiple_assign_taxonomy_output/"
base_out_dir = "/home/caporaso/analysis/short-read-tax-assignment/data/eval-pre-computed/"

gregcaporaso

#!/usr/bin/env python

from os.path import join

query_fp = "/home/caporaso/analysis/short-read-tax-assignment/data/qiime-mock-community/S16S-2/rep_set.fna"
reference_seqs_fp = "/data/gg_13_5_otus/rep_set/97_otus.fasta"
reference_tax_fp = "/data/gg_13_5_otus/taxonomy/97_otu_taxonomy.txt"
input_biom_fp = "/home/caporaso/analysis/short-read-tax-assignment/data/qiime-mock-community/S16S-2/otu_table_mc2_no_pynast_failures.biom"
output_biom_fn = "otu_table_mc2_no_pynast_failures_w_taxa.biom"
output_dir = "/home/caporaso/analysis/short-read-tax-assignment/demo/eval-demo/usearch_v_97/"

gregcaporaso

pick_otus:enable_rev_strand_match True
pick_otus:max_accepts 1
pick_otus:max_rejects 8
pick_otus:stepwords 8
pick_otus:word_length 8