sampledoc
News and Announcements »

parallel_multiple_rarefactions.py – Parallel multiple file rarefaction

Description:

This script performs like the multiple_rarefactions.py script, but is intended to make use of multicore/multiprocessor environments to perform analyses in parallel.

Usage: parallel_multiple_rarefactions.py [options]

Input Arguments:

Note

[REQUIRED]

-i, --input_path
Input filepath, (the otu table) [REQUIRED]
-o, --output_path
Write output rarefied otu tables here makes dir if it doesn’t exist [REQUIRED]
-m, --min
Min seqs/sample [REQUIRED]
-x, --max
Max seqs/sample (inclusive) [REQUIRED]

[OPTIONAL]

-n, --num-reps
Num iterations at each seqs/sample level [default: 10]
--suppress_lineages_included
Exclude taxonomic (lineage) information for each OTU.
-s, --step
Levels: min, min+step... for level <= max [default: 1]
--subsample_multinomial
Subsample using subsampling with replacement [default: False]
-R, --retain_temp_files
Retain temporary files after runs complete (useful for debugging) [default: False]
-S, --suppress_submit_jobs
Only split input and write commands file - don’t submit jobs [default: False]
-T, --poll_directly
Poll directly for job completion rather than running poller as a separate job. If -T is specified this script will not return until all jobs have completed. [default: False]
-U, --cluster_jobs_fp
Path to cluster jobs script (defined in qiime_config) [default: start_parallel_jobs.py]
-W, --suppress_polling
Suppress polling of jobs and merging of results upon completion [default: False]
-X, --job_prefix
Job prefix [default: descriptive prefix + random chars]
-Z, --seconds_to_sleep
Number of seconds to sleep between checks for run completion when polling runs [default: 1]
-O, --jobs_to_start
Number of jobs to start [default: 2]

Output:

The result of parallel_multiple_rarefactions.py consists of a number of files, which depend on the minimum/maximum number of sequences per samples, steps and iterations. The files have the same otu table format as the input otu_table.biom, and are named in the following way: rarefaction_100_0.txt, where “100” corresponds to the sequences per sample and “0” for the iteration.

OTU tables of different depths:

Build rarefied otu tables containing 10 (-m) to 140 (-x) sequences in steps of 10 (-s) with 2 (-n) repetions per number of sequences, from otu_table.biom (-i). Write the output files to the rarefied_otu_tables directory (-o, will be created if it doesn’t exist). The name of the output files will be of the form rarefaction_<num_seqs>_<reptition_number>.biom. ALWAYS SPECIFY ABSOLUTE FILE PATHS (absolute path represented here as $PWD, but will generally look something like /home/ubuntu/my_analysis/).

parallel_multiple_rarefactions.py -o $PWD/rarefied_otu_tables/ -m 10 -x 140 -s 10 -n 2 -i $PWD/otu_table.biom

OTU tables of the same depth:

Build 8 rarefied otu tables each containing exactly 100 sequences per sample (even depth rarefaction). ALWAYS SPECIFY ABSOLUTE FILE PATHS (absolute path represented here as $PWD, but will generally look something like /home/ubuntu/my_analysis/).

parallel_multiple_rarefactions.py -o $PWD/even_otu_tables/ -m 100 -x 100 -n 8 -i $PWD/otu_table.biom

Site index


sampledoc