Source code for galaxy.datatypes.converters.pbed_ldreduced_converter

# converter for ldreduced rgenetics datatype
# used for grr and eigenstrat - shellfish if we get around to it

import os
import subprocess
import sys
import tempfile
import time

prog = ""

galhtmlprefix = """<?xml version="1.0" encoding="utf-8" ?>
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "">
<html xmlns="" xml:lang="en" lang="en">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
<meta name="generator" content="Galaxy %s tool output - see" />
<link rel="stylesheet" href="/static/style/base.css" type="text/css" />
<div class="document">

plinke = 'plink'

[docs]def timenow(): """return current time as a string """ return time.strftime('%d/%m/%Y %H:%M:%S', time.localtime(time.time()))
[docs]def pruneLD(plinktasks=[], cd='./', vclbase=[]): """ """ fd, plog = tempfile.mkstemp() os.close(fd) alog = [] alog.append('## Rgenetics: Galaxy Tools Plink pruneLD runner\n') for task in plinktasks: # each is a list vcl = vclbase + task with open(plog, 'w') as sto: subprocess.check_call(vcl, stdout=sto, stderr=sto, cwd=cd) try: lplog = open(plog).readlines() lplog = [elem for elem in lplog if elem.find('Pruning SNP') == -1] alog += lplog alog.append('\n') os.unlink(plog) # no longer needed except Exception: alog.append('### {} Strange - no std out from plink when running command line\n{}\n'.format(timenow(), ' '.join(vcl))) return alog
[docs]def makeLDreduced(basename, infpath=None, outfpath=None, plinke='plink', forcerebuild=False, returnFname=False, winsize="60", winmove="40", r2thresh="0.1"): """ not there so make and leave in output dir for post job hook to copy back into input extra files path for next time """ outbase = os.path.join(outfpath, basename) inbase = os.path.join(infpath) plinktasks = [] vclbase = [plinke, '--noweb'] plinktasks += [['--bfile', inbase, '--indep-pairwise {} {} {}'.format(winsize, winmove, r2thresh), '--out %s' % outbase], ['--bfile', inbase, '--extract {} --make-bed --out {}'.format(outbase, outbase)]] vclbase = [plinke, '--noweb'] pruneLD(plinktasks=plinktasks, cd=outfpath, vclbase=vclbase)
[docs]def main(): """ need to work with rgenetics composite datatypes so in and out are html files with data in extrafiles path .. raw:: xml <command> python '$__tool_directory__/' '$input1.extra_files_path/$input1.metadata.base_name' '$winsize' '$winmove' '$r2thresh' '$output1' '$output1.files_path' 'plink' </command> """ nparm = 7 if len(sys.argv) < nparm: sys.stderr.write('## %s called with %s - needs %d parameters \n' % (prog, sys.argv, nparm)) sys.exit(1) inpedfilepath = sys.argv[1] base_name = os.path.split(inpedfilepath)[-1] winsize = sys.argv[2] winmove = sys.argv[3] r2thresh = sys.argv[4] outhtmlname = sys.argv[5] outfilepath = sys.argv[6] try: os.makedirs(outfilepath) except Exception: pass plink = sys.argv[7] makeLDreduced(base_name, infpath=inpedfilepath, outfpath=outfilepath, plinke=plink, forcerebuild=False, returnFname=False, winsize=winsize, winmove=winmove, r2thresh=r2thresh) flist = os.listdir(outfilepath) with open(outhtmlname, 'w') as f: f.write(galhtmlprefix % prog) s1 = '## Rgenetics: Galaxy Tools {} {}'.format(prog, timenow()) # becomes info s2 = 'Input {}, winsize={}, winmove={}, r2thresh={}'.format(base_name, winsize, winmove, r2thresh) print('{} {}'.format(s1, s2)) f.write('<div>{}\n{}\n<ol>'.format(s1, s2)) for i, data in enumerate(flist): f.write('<li><a href="{}">{}</a></li>\n'.format(os.path.split(data)[-1], os.path.split(data)[-1])) f.write("</div></body></html>")
if __name__ == "__main__": main()