High-Performance Computing at the NIH
GitHub YouTube @nih_hpc RSS Feed
DANPOS

A toolkit for Dynamic Analysis of Nucleosome and Protein Occupancy by Sequencing

Web site
Reference

DANPOS On Helix
back to top

The author has provided test data which can be found at /usr/local/apps/danpos/2.2.2/testdata. These data can be read from the directory for testing without actually being copied as in the following example. (User input in bold.)

[user@helix ~]$ mkdir -p /data/$USER/danpos_test

[user@helix ~]$ cd /data/$USER/danpos_test

[user@helix danpos_test]$ module load danpos
[+] Loading danpos 2.2.2 on helix.nih.gov
[+] Loading python 2.7.10 ...
[+] Loading rpy2 2.7.0 (R version 3.2.2) ...
[+] Loading samtools 1.3.1 ...

[user@helix danpos_test]$ danpos.py -h
danpos version 2.2.2
For help information for each function, try:
python danpos.py  -h

Functions:
	dpos:
		analyze each protein-binding position (~100
		to ~200bp wide) across the whole genome,
		e.g. nucleosome positions.
	dpeak:
		analyze each protein-binding peak (~1 to ~1kb
		wide) across the whole genome, e.g. protein
		that binds accruately to some specific motifs.
	dregion:
		analyze each protein-binding region (~1 to
		~10kb wide) across the whole genome, e.g.
		some histone modifications.
	dtriple:
		Do analysis at all three levels including each
		region, peak, and position. Would be useful
		when little is known about the potential binding
		pattern.
	profile:
		analyze wiggle format occupancy or differential
		signal profile relative to gene structures or
		bed format genomic regions.
	wiq:
		normalize wiggle files to have the same quantiles (Quantile normalization).
	wig2wiq:
		convert wiggle format file to wiq format.
	stat:
		some statistics for positions, peaks, or regions.
	selector:
		select a subset of positions, peaks, or regions
		by value ranges or gene structures neighboring.
	valuesAtRanks:
		retrieve position, peak, or region values by ranks.

Kaifu Chen, et al. chenkaifu@gmail.com, Li lab, Biostatistics department, Dan L. Duncan cancer center, Baylor College of Medicine.

[user@helix danpos_test]$ TEST_DATA=/usr/local/apps/danpos/2.2.2/testdata 

[user@helix danpos_test]$ danpos.py dpos $TEST_DATA/nucleosome_sampleA_rep1.bowtie > danpos_dpos.log

[user@helix danpos_test]$ ls -l
total 12
-rw-r----- 1 user group 6338 Jul 25 14:10 danpos_dpos.log
drwxr-x--- 3 user group 4096 Jul 25 14:08 result
To see the contents of danpos_dpos.log click here.

Running a single DANPOS job on Biowulf
back to top

Set up a batch script along the following lines:

#!/bin/bash
# file called myjob.bat

module load danpos
cd /data/$USER/output_dir
TEST_DATA=/usr/local/apps/danpos/2.2.2/testdata
danpos.py dpos $TEST_DATA/nucleosome_sampleA_rep1.bowtie
danpos.py dpos $TEST_DATA/nucleosome_sampleA_rep1.bowtie:$TEST_DATA/nucleosome_sampleB.bowtie

Submit this job with:

[user@biowulf ~]$ sbatch myjob.bat

For more information on submitting jobs to slurm, see Job Submission in the Biowulf User Guide.

Running a swarm of DANPOS jobs on Biowulf
back to top

Sample swarm command file

# --------file myjobs.swarm----------
danpos.py dpos /dir/to/data/rep1.bowtie 
danpos.py dpos /dir/to/data/rep2.bowtie 
danpos.py dpos /dir/to/data/rep3.bowtie 
....
danpos.py dpos /dir/to/data/repN.bowtie 
# -----------------------------------

Submit this set of runs to the batch system by typing

[user@biowulf ~]$ swarm --module danpos -f myjobs.swarm

For details on using swarm see Swarm on Biowulf.

Documentation
back to top

DANPOS is extensively documented. To read the help doc, type danpos.py -h. You can also view information about specific commands. For instance, danpos.py dpos -h returns detailed help about the dpos command. See also: