# working directories
chunk_dir = wd + '/chunks_for_imp'
imp_dir = wd + '/imp_sub'
os.mkdir(chunk_dir)
os.mkdir(imp_dir)
print 'Genomic chunks: %s' % chunk_dir
print 'Imputation: %s' % imp_dir
######################
print '\n...Creating genomic chunks for imputation...'
######################
os.chdir(chunk_dir)
link(str(shape_dir)+'/'+str(args.bfile)+'.hg19.ch.fl.bim', str(args.bfile)+'.hg19.ch.fl.bim', 'reference-aligned bim file')
# create chunks
chunk_call = [chunker_ex,
'--bfile',str(args.bfile)+'.hg19.ch.fl',
'--out',str(args.out),
addout_txt[0],addout_txt[1],
'--Mb-size',str(args.Mb_size),
'--snp-size',str(args.snp_size),
'--chr-info-file',str(args.chr_info_file)]
chunk_call = filter(None,chunk_call)
chunk_log = open('chunk.'+str(outdot)+'.log', 'w')
print ' '.join(chunk_call) + '\n'
subprocess.check_call(chunk_call, stderr=subprocess.STDOUT, stdout=chunk_log)
chunk_log.close()
fam.close()
outfam.close()
famtrans.close()
print 'New fam file written to: %s\n' % (str(pi_dir) + '/' + str(outfam_file))
print 'Fam file translation written to: %s\n' % (str(pi_dir) + '/' +
str(famtrans_file))
######################
print '\n...Splitting by chromosome...'
######################
os.chdir(wd)
os.mkdir('phase_chr')
os.chdir(wd + '/phase_chr')
link(pi_dir + '/' + str(args.bfile) + '.hg19.ch.fl.bed',
str(args.bfile) + '.hg19.ch.fl.bed', 'aligned .bed file')
link(pi_dir + '/' + str(args.bfile) + '.hg19.ch.fl.bim',
str(args.bfile) + '.hg19.ch.fl.bim', 'aligned .bim file')
link(pi_dir + '/' + str(args.bfile) + '.hg19.ch.fl.fam.idnum',
str(args.bfile) + '.hg19.ch.fl.fam', 'aligned .fam file')
link(pi_dir + '/' + str(args.bfile) + '.hg19.ch.fl.fam.transl',
str(args.bfile) + '.hg19.ch.fl.fam.transl', 'fam number translation file')
# TODO: handle empty chromosomes
for i in xrange(1, 23):
chr_log = open(str(outdot) + '.chr' + str(i) + '.log', 'w')
chr_call = [
plinkx, '--bfile',
str(args.bfile) + '.hg19.ch.fl', '--chr',
str(i), '--make-bed', '--silent', '--memory',
str(2000), '--out',
print '\n############'
print '\n'
print 'All jobs submitted.\n'
exit(0)
######################
print '\n...Setting up working directory...'
######################
# working directories
proc_dir = wd + '/imp_postproc'
os.mkdir(proc_dir)
print 'Impute2 output processing: %s' % proc_dir
os.chdir(proc_dir)
link(
str(chunk_dir) + '/' + str(outdot) + '.chunks.txt',
str(outdot) + '.chunks.txt', 'genomic chunk results')
######################
print '\n...Generating best-guess genotypes...'
######################
# TODO: flex queue/mem reqs
uger_bg_template = """#!/usr/bin/env sh
#$ -j y
#$ -cwd
#$ -V
#$ -N {jname}
#$ -q short
#$ -l m_mem_free=4g
#$ -t 1-{nchunk}
#############
qcdir = 'qc_'+str(args.out)
print '\n...Setting up working directory (./%s)...' % qcdir
#############
wd = os.getcwd()
if os.path.isdir(qcdir):
raise IOError ('Output directory %s already exists. Stopping to prevent overwriting files.' % qcdir)
else:
os.makedirs(qcdir)
os.chdir(qcdir)
# link plink files (with verification)
link(str(wd+'/'+args.bfile+'.bed'), str(args.bfile+'.bed'), 'input bed file')
link(str(wd+'/'+args.bfile+'.bim'), str(args.bfile+'.bim'), 'input bim file')
if args.skip_fid_tags:
link(str(wd+'/'+args.bfile+'.fam'), str(args.bfile+'.fam'), 'input fam file')
else:
link(str(wd+'/'+args.bfile+'.fam'), str(args.bfile+'.fam.original'), 'input fam file')
# setup SNP, ID QC fail lists
snpout_nam = args.out + '.exclude_snps.txt'
idout_nam = args.out + '.exclude_iids.txt'
#############
if not args.skip_fid_tags:
print '\n...Preparing tags for FIDs...'
fam.close()
outfam.close()
famtrans.close()
print 'New fam file written to: %s\n' % (str(pi_dir)+'/'+str(outfam_file))
print 'Fam file translation written to: %s\n' % (str(pi_dir)+'/'+str(famtrans_file))
######################
print '\n...Splitting by chromosome...'
######################
os.chdir(wd)
os.mkdir('phase_chr')
os.chdir(wd+'/phase_chr')
link(pi_dir + '/' +str(args.bfile) +'.hg19.ch.fl.bed', str(args.bfile) +'.hg19.ch.fl.bed', 'aligned .bed file')
link(pi_dir + '/' +str(args.bfile) +'.hg19.ch.fl.bim', str(args.bfile) +'.hg19.ch.fl.bim', 'aligned .bim file')
link(pi_dir + '/' +str(args.bfile) +'.hg19.ch.fl.fam.idnum', str(args.bfile) +'.hg19.ch.fl.fam', 'aligned .fam file')
link(pi_dir + '/' +str(args.bfile) +'.hg19.ch.fl.fam.transl', str(args.bfile) +'.hg19.ch.fl.fam.transl', 'fam number translation file')
# TODO: handle empty chromosomes
for i in xrange(1,23):
chr_log = open(str(outdot) + '.chr' + str(i) + '.log', 'w')
chr_call = [plinkx,
'--bfile', str(args.bfile) + '.hg19.ch.fl',
'--chr', str(i),
'--make-bed',
'--silent',
'--memory', str(2000),
'--out', str(args.bfile) + '.hg19.ch.fl.chr' + str(i)]
print 'Begin!'
print '############'
#############
print '\n...Setting up working directory (%s)...' % str(args.outdir)
#############
wd = os.getcwd()
if not os.path.isdir(str(args.outdir)):
os.makedirs(str(args.outdir))
os.chdir(args.outdir)
# link plink files (with verification)
link(str(wd + '/' + args.unrel_bfile + '.bed'), str(args.unrel_bfile + '.bed'),
'bed file for unrelated individuals')
link(str(wd + '/' + args.unrel_bfile + '.bim'), str(args.unrel_bfile + '.bim'),
'bim file for unrelated individuals')
link(str(wd + '/' + args.unrel_bfile + '.fam'), str(args.unrel_bfile + '.fam'),
'fam file for unrelated individuals')
link(str(wd + '/' + args.target_bfile + '.bed'),
str(args.target_bfile + '.bed'), 'bed file for target individuals')
link(str(wd + '/' + args.target_bfile + '.bim'),
str(args.target_bfile + '.bim'), 'bim file for target individuals')
link(str(wd + '/' + args.target_bfile + '.fam'),
str(args.target_bfile + '.fam'), 'fam file for target individuals')
# link pca file, if provided
if not (args.plot_admix_pca == None or args.plot_admix_pca == "None"):
#############
qcdir = 'qc_'+str(args.out)
print '\n...Setting up working directory (./%s)...' % qcdir
#############
wd = os.getcwd()
if os.path.isdir(qcdir):
raise IOError ('Output directory %s already exists. Stopping to prevent overwriting files.' % qcdir)
else:
os.makedirs(qcdir)
os.chdir(qcdir)
# link plink files (with verification)
link(str(wd+'/'+args.bfile+'.bed'), str(args.bfile+'.bed'), 'input bed file')
link(str(wd+'/'+args.bfile+'.bim'), str(args.bfile+'.bim'), 'input bim file')
if args.skip_fid_tags:
link(str(wd+'/'+args.bfile+'.fam'), str(args.bfile+'.fam'), 'input fam file')
else:
link(str(wd+'/'+args.bfile+'.fam'), str(args.bfile+'.fam.original'), 'input fam file')
# setup SNP, ID QC fail lists
snpout_nam = args.out + '.exclude_snps.txt'
idout_nam = args.out + '.exclude_iids.txt'
#############
if not args.skip_fid_tags:
print '\n...Preparing tags for FIDs...'
print '\n'
print '############'
print 'Begin!'
print '############'
######################
print '\n...Setting up working directory ./%s...' % str(outdir)
######################
wd = os.getcwd()
os.mkdir(outdir)
os.chdir(outdir)
link(wd+'/'+str(args.bfile)+'.bed', str(args.bfile)+'.bed', 'input plink bed file')
link(wd+'/'+str(args.bfile)+'.bim', str(args.bfile)+'.bim', 'input plink bim file')
link(wd+'/'+str(args.bfile)+'.fam', str(args.bfile)+'.fam', 'input plink fam file')
if str(args.strict_bfile) != str(args.bfile):
link(wd+'/'+str(args.strict_bfile)+'.bed', str(args.strict_bfile)+'.bed', 'plink bed file for GRM')
link(wd+'/'+str(args.strict_bfile)+'.bim', str(args.strict_bfile)+'.bim', 'plink bim file for GRM')
link(wd+'/'+str(args.strict_bfile)+'.fam', str(args.strict_bfile)+'.fam', 'plink fam file for GRM')
if args.covar is not None and str(args.covar) != "None":
link(wd+'/'+str(args.covar), str(args.covar), 'covariate file')
if args.pheno is not None and str(args.pheno) != "None":
link(wd+'/'+str(args.pheno), str(args.pheno), 'phenotype file')
if args.info_file is not None and str(args.info_file) != "None":
# working directories
chunk_dir = wd + '/chunks_for_imp'
imp_dir = wd + '/imp_sub'
os.mkdir(chunk_dir)
os.mkdir(imp_dir)
print 'Genomic chunks: %s' % chunk_dir
print 'Imputation: %s' % imp_dir
######################
print '\n...Creating genomic chunks for imputation...'
######################
os.chdir(chunk_dir)
link(str(shape_dir)+'/'+str(args.bfile)+'.hg19.ch.fl.bim', str(args.bfile)+'.hg19.ch.fl.bim', 'reference-aligned bim file')
# create chunks
chunk_call = [chunker_ex,
'--bfile',str(args.bfile)+'.hg19.ch.fl',
'--out',str(args.out),
addout_txt[0],addout_txt[1],
'--Mb-size',str(args.Mb_size),
'--snp-size',str(args.snp_size),
'--chr-info-file',str(args.chr_info_file)]
chunk_call = filter(None,chunk_call)
chunk_log = open('chunk.'+str(outdot)+'.log', 'w')
print ' '.join(chunk_call) + '\n'
subprocess.check_call(chunk_call, stderr=subprocess.STDOUT, stdout=chunk_log)
chunk_log.close()
print '\n'
print 'All jobs submitted.\n'
exit(0)
######################
print '\n...Setting up working directory...'
######################
# working directories
proc_dir = wd + '/imp_postproc'
os.mkdir(proc_dir)
print 'Impute2 output processing: %s' % proc_dir
os.chdir(proc_dir)
link(str(chunk_dir)+'/'+str(outdot)+'.chunks.txt', str(outdot)+'.chunks.txt', 'genomic chunk results')
######################
print '\n...Generating best-guess genotypes...'
######################
# TODO: flex queue/mem reqs
uger_bg_template = """#!/usr/bin/env sh
#$ -j y
#$ -cwd
#$ -V
#$ -N {jname}
#$ -q short
#$ -l m_mem_free=4g
#$ -t 1-{nchunk}
print '############'
#############
print '\n...Setting up working directory (%s)...' % str(args.outdir)
#############
wd = os.getcwd()
if not os.path.isdir(str(args.outdir)):
os.makedirs(str(args.outdir))
os.chdir(args.outdir)
# link plink files (with verification)
if run_admix or args.use_exemplars:
link(str(wd + '/' + args.unrel_bfile + '.bed'),
str(args.unrel_bfile + '.bed'), 'bed file for unrelated individuals')
link(str(wd + '/' + args.unrel_bfile + '.bim'),
str(args.unrel_bfile + '.bim'), 'bim file for unrelated individuals')
link(str(wd + '/' + args.unrel_bfile + '.fam'),
str(args.unrel_bfile + '.fam'), 'fam file for unrelated individuals')
link(str(wd + '/' + args.target_bfile + '.bed'),
str(args.target_bfile + '.bed'), 'bed file for target individuals')
link(str(wd + '/' + args.target_bfile + '.bim'),
str(args.target_bfile + '.bim'), 'bim file for target individuals')
link(str(wd + '/' + args.target_bfile + '.fam'),
str(args.target_bfile + '.fam'), 'fam file for target individuals')
# link pca file, if provided
if not (args.plot_admix_pca == None or args.plot_admix_pca == "None"):
link(os.path.normpath(str(wd + '/' + args.plot_admix_pca)),
print '\n'
print '############'
print 'Begin!'
print '############'
######################
print '\n...Setting up working directory ./%s...' % str(outdir)
######################
wd = os.getcwd()
os.mkdir(outdir)
os.chdir(outdir)
link(wd+'/'+str(args.bfile)+'.bed', str(args.bfile)+'.bed', 'input plink bed file')
link(wd+'/'+str(args.bfile)+'.bim', str(args.bfile)+'.bim', 'input plink bim file')
link(wd+'/'+str(args.bfile)+'.fam', str(args.bfile)+'.fam', 'input plink fam file')
if str(args.strict_bfile) != str(args.bfile):
link(wd+'/'+str(args.strict_bfile)+'.bed', str(args.strict_bfile)+'.bed', 'plink bed file for GRM')
link(wd+'/'+str(args.strict_bfile)+'.bim', str(args.strict_bfile)+'.bim', 'plink bim file for GRM')
link(wd+'/'+str(args.strict_bfile)+'.fam', str(args.strict_bfile)+'.fam', 'plink fam file for GRM')
if args.covar is not None and str(args.covar) != "None":
link(wd+'/'+str(args.covar), str(args.covar), 'covariate file')
if args.pheno is not None and str(args.pheno) != "None":
link(wd+'/'+str(args.pheno), str(args.pheno), 'phenotype file')
if args.info_file is not None and str(args.info_file) != "None":
print 'Begin!'
print '############'
#############
print '\n...Setting up working directory (%s)...' % str(args.outdir)
#############
wd = os.getcwd()
if not os.path.isdir(str(args.outdir)):
os.makedirs(str(args.outdir))
os.chdir(args.outdir)
# link plink files (with verification)
link(str(wd+'/'+args.unrel_bfile+'.bed'), str(args.unrel_bfile+'.bed'), 'bed file for unrelated individuals')
link(str(wd+'/'+args.unrel_bfile+'.bim'), str(args.unrel_bfile+'.bim'), 'bim file for unrelated individuals')
link(str(wd+'/'+args.unrel_bfile+'.fam'), str(args.unrel_bfile+'.fam'), 'fam file for unrelated individuals')
link(str(wd+'/'+args.target_bfile+'.bed'), str(args.target_bfile+'.bed'), 'bed file for target individuals')
link(str(wd+'/'+args.target_bfile+'.bim'), str(args.target_bfile+'.bim'), 'bim file for target individuals')
link(str(wd+'/'+args.target_bfile+'.fam'), str(args.target_bfile+'.fam'), 'fam file for target individuals')
# link pca file, if provided
if not (args.plot_admix_pca==None or args.plot_admix_pca=="None"):
link(str(wd+'/'+args.plot_admix_pca), str(args.plot_admix_pca), 'PCA file')
#############
