dumphead = chunks_in.readline()
for line in chunks_in:
(chrom, start, end, chname) = line.split()
chunks[str(chname)] = [str(chrom), int(start), int(end)]
# verify output file exists
ch_imp = imp_dir + '/' + str(outdot) + '.imp.' + str(chname) + '.gz'
# verify completed successfully
# - based on expected output of concordance table on last line
ch_sum = imp_dir + '/' + str(outdot) + '.imp.' + str(chname) + '_summary'
# record failed chunks
if not os.path.isfile(ch_imp):
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
elif '[0.9-1.0]' not in file_tail(ch_sum, n=1):
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
chunks_in.close()
###############
# if there are missing chunks, restart their imputation and resub agg script
###############
if len(mis_chunks) > 0:
nummiss = len(mis_chunks)
print 'Missing results for %d imputation chunks. Preparing to resubmit...' % nummiss
# check if already tried with this number of missing chunks
os.chdir(imp_dir)
tmp_chunk_file_name = 'tmp_missing_' + str(nummiss) + '_chunks.' + str(
elif args.model == 'logistic':
ch_out = 'logis.'+str(outdot)+'.'+str(chname)+'.assoc.logistic'
out_len = 12
elif args.model == 'linear':
ch_out = 'linear.'+str(outdot)+'.'+str(chname)+'.assoc.linear'
out_len = 12
# record chunks with no/partial/broken output
if not os.path.isfile(ch_out):
print 'Output not found for %s' % str(ch_out)
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
elif file_len(ch_out) < file_len(str(outdot)+'.snps.'+str(chname)+'.txt'):
print 'Output file %s is incomplete' % str(ch_out)
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
else:
ft = file_tail(ch_out)
if len(ft.split()) != out_len:
print 'Last line of output file %s is incomplete' % str(ch_out)
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
chunks_in.close()
###############
# if there are missing chunks, restart their gwas and resub agg script
###############
if len(mis_chunks) > 0:
nummiss = len(mis_chunks)
print '\nMissing results for %d GWAS jobs. Preparing to resubmit...' % nummiss
# just missing chunks for task array
ch_out = 'dfam.'+str(outdot)+'.'+str(chname)+'.dfam'
out_len = 8
elif args.model == 'gmmat':
ch_out = 'gmmat_score.'+str(outdot)+'.'+str(chname)+'.R.txt'
out_len = 11
elif args.model == 'gmmat-fam':
ch_out = 'gmmatfam_score.'+str(outdot)+'.'+str(chname)+'.R.txt'
out_len = 11
# record chunks with no/partial/broken output
if not os.path.isfile(ch_out):
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
elif file_len(ch_out) != file_len(str(outdot)+'.snps.'+str(chname)+'.txt'):
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
else:
ft = file_tail(ch_out)
if len(ft.split()) != out_len:
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
chunks_in.close()
###############
# if there are missing chunks, restart their gwas and resub agg script
###############
if len(mis_chunks) > 0:
nummiss = len(mis_chunks)
print 'Missing results for %d GWAS jobs. Preparing to resubmit...' % nummiss
# just missing chunks for task array
# fail if already tried
dumphead = chunks_in.readline()
for line in chunks_in:
(chrom, start, end, chname) = line.split()
chunks[str(chname)] = [str(chrom), int(start), int(end)]
# verify output file exists
ch_imp = imp_dir + '/' + str(outdot) + '.imp.' + str(chname) + '.gz'
# verify completed successfully
# - based on expected output of concordance table on last line
ch_sum = imp_dir + '/' + str(outdot) + '.imp.' + str(chname) + '_summary'
# record failed chunks
if not os.path.isfile(ch_imp):
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
elif '[0.9-1.0]' not in file_tail(ch_sum, n=1):
mis_chunks[str(chname)] = [str(chrom), int(start), int(end)]
chunks_in.close()
###############
# if there are missing chunks, restart their imputation and resub agg script
###############
if len(mis_chunks) > 0:
nummiss = len(mis_chunks)
print 'Missing results for %d imputation chunks. Preparing to resubmit...' % nummiss
# check if already tried with this number of missing chunks