def munge_transcript(data, RefSeqs):
"""
Return HGVS correct transcript annotations
Filtered with a preferred transcript list
NM_006772.1:c.1713G>A
"""
codon, prot, protein, coding, txpt = ' ', ' ', ' ', ' ', None
transcripts = data.get('Transcripts')
if transcripts is not None:
# Split incoming trans, strip the trailing )
data = multi_split(transcripts.replace('(', ':'), ',(')
for d in data:
# Split the actual transcript info which is colon separated
x = d.split(':')
try:
gene, txpt, exon, codon, prot = x
except ValueError:
try:
gene, txpt, exon, codon = x
except ValueError:
continue
pref_trans = RefSeqs.get(txpt)
#Want to return None for all values if not pref_trans
if not pref_trans:
continue
code = pref_trans + ':' + codon
coding = coding + ' ' + code
protein = protein + ' ' + prot
return coding.strip(), protein.strip()
def parse_sv_alt(data):
''' Split the ALT field from breakend format
]7:55248960]A to chr7:55248960
A[7:55249011[ to chr7:55249011
'''
if data['ID'].endswith(('o', 'h')):
a, b = multi_split(data['ALT'], '[]')
#Create Event2
#the position has a : in it while the sequence does not
#Only process chr1-23, X, Y
if ':' in a:
try:
chrom = a.split(':')
data['Event2'] = 'chr' + str(chromosomes[a[0]]) + str(a[1:])
data['Seq'] = b
except KeyError:
pass
else:
try:
chrom = b.split(':')
data['Event2'] = 'chr' + str(chromosomes[b[0]]) + str(b[1:])
data['Seq'] = a
except KeyError:
pass
elif data['ID'].endswith('b'):
data['Event2'] = 'SingleBreakEnd'
data['Seq'] = data['ALT']
else:
print("some weird data was encountered", data)
sys.exit()
return pd.Series(data)
def process_files(infiles, tab, filetype):
"""
Rename the analysis files for workbook
"""
for fname in infiles:
(f_path, f_name) = os.path.split(fname)
if re.search(str(filetype), f_name):
(f_short_name, f_extension) = os.path.splitext(f_name)
sheet_name = multi_split(f_short_name, '._')
try:
#48_A03_BROv7_HA0186_NA12878_QC_Analysis
if sheet_name[-2] == 'QC':
sheet_name = '0_QC'
#48_A03_BROv7_HA0186_NA12878_Quality_Analysis
elif sheet_name[-2] == 'Quality':
sheet_name = '1_QC_Metrics'
#48_A03_BROv7_HA0186_NA12878_CNV_QC_[Exon/Gene]_Analysis
elif sheet_name[-3] == 'QC':
if sheet_name[-2] == 'Gene':
sheet_name = '2_QC_by_Gene'
elif sheet_name[-2] == 'Exon':
sheet_name = '3_QC_by_Exon'
#48_A03_BROv7_HA0186_NA12878_CNV_[Exon/Gene]_Analysis
elif sheet_name[-3] == 'CNV':
if sheet_name[-2] == 'Gene':
sheet_name = '7_CNV_Gene'
elif sheet_name[-2] == 'Exon':
sheet_name = '8_CNV_Exon'
#48_A03_BROv7_HA0186_NA12878_SV_Analysis
elif sheet_name[-2] == 'SV':
sheet_name = '4_SV_Gridss'
#48_A03_BROv7_HA0186_NA12878_Breakdancer_Analysis
elif sheet_name[-2] == 'Breakdancer':
sheet_name = '5_SV_Breakdancer'
#48_A03_BROv7_HA0186_NA12878_Pindel_Analysis
elif sheet_name[-2] == 'Pindel':
sheet_name = '6_SV_Pindel'
#48_A03_BROv7_HA0186_NA12878_Genotype_Analysis
elif sheet_name[-2] == 'Genotype':
sheet_name = '9_Clinically_Flagged'
#48_A03_BROv7_HA0186_NA12878_MSI_Analysis
elif sheet_name[-2] == 'MSI':
sheet_name = '12_MSI'
#48_A03_BROv7_HA0186_NA12878_Amplicon_Analysis
elif sheet_name[-2] == 'Amplicon':
sheet_name = '13_Amplicons'
#48_A03_BROv7_HA0186_NA12878_PolyHunter_Analysis
elif sheet_name[-2] == 'PolyHunter':
sheet_name = '14_PolyHunter'
#48_A03_BROv7_HA0186_NA12878_SNP_Indel_Analysis
elif sheet_name[-2] == 'SNP':
sheet_name = '10_SNP'
#48_A03_BROv7_HA0186_NA12878_INDEL_Analysis
elif sheet_name[-2] == 'INDEL':
sheet_name = '11_INDEL'
if sheet_name == tab:
return sheet_name, fname
except IndexError:
continue
def testSplitString(self):
"""
Tests spliting a string given string of split characters
Used to split path names in crawlers
"""
result=multi_split('/home/genetics/data/run_info', '/_')
self.assertEquals(result, ['home','genetics','data','run','info'])
def testSplitString(self):
"""
Tests spliting a string given string of split characters
Used to split path names in crawlers
"""
result = ann.multi_split('/home/genetics/data/run_info', '/_')
self.assertEquals(result, ['home', 'genetics', 'data', 'run', 'info'])
def munge_pfx(pfx):
"""
Change the pfx output in files to a shorter version
"""
output=multi_split(pfx, '/_.')
keys=['run','well','library-version','machine-run','control']
pfx_info=dict(zip(keys,output))
pfx_info['control']=check_control(pfx_info['control'])
pfx_info['mini-pfx']='{well}{control}'.format(**pfx_info)
pfx_info['pfx']='{well}{control}_{library-version}'.format(**pfx_info)
return pfx_info
def munge_path(pth):
"""
Get date, run, project, machine, assay, prep-type from path
"""
output = multi_split(pth, '/_')
#Assuming we want YYMMDD_RUN_PROJECT
if output[-1] == 'output':
output = output[-4:-1]
keys = ['date', 'run', 'project']
#If old version of data that isn't in a 'output' subfolder
elif len(output) == 5:
keys = ['date', 'machine', 'assay', 'run', 'version', 'project']
#check that the third item is the assay
if not output[2].lower() in ASSAY_CODES.values():
#is the second item is the assay?
if not output[1].lower() in ASSAY_CODES.values():
print "not a good path", output
project = output[1] + output[3].strip('run')
output = [
output[0], output[2], output[1], output[3], output[4], project
]
else:
project = output[2] + output[3].strip('run')
output = [
output[0], output[1], output[2], output[3], output[4], project
]
else:
output = output[-3:]
keys = ['date', 'run', 'project']
pathinfo = dict(zip(keys, output))
pathinfo['date'] = munge_date(pathinfo['date'])
pathinfo['project'] = pathinfo['project'].lower()
#Set Machine
if re.search('[%s][A-Z]\d+' % ''.join(MACHINE_CODES.keys()),
pathinfo['run']):
pathinfo['machine'] = MACHINE_CODES[pathinfo['run'][0]]
#Set assay
for a in ASSAY_CODES.keys():
if re.search(a, pathinfo['project'], re.IGNORECASE):
pathinfo['assay'] = ASSAY_CODES[a]
elif re.search('MSI', pathinfo['project'], re.IGNORECASE):
pathinfo['assay'] = 'msi-plus'
#Set prep type
if re.search('kapa', pathinfo['project']):
pathinfo['prep_type'] = 'kapa'
elif re.search('hotspot', pathinfo['assay']):
pathinfo['prep_type'] = 'truseq'
else:
pathinfo['prep_type'] = 'sure_select'
return pathinfo
def munge_path(pth):
"""
Get date, run, project, machine, assay, prep-type from path
"""
output=multi_split(pth, '/_')
#Assuming we want YYMMDD_RUN_PROJECT
if output[-1]=='output':
output=output[-4:-1]
keys=['date','run', 'project']
#If old version of data that isn't in a 'output' subfolder
elif len(output)==5:
keys=['date','machine','assay','run','version','project']
#check that the third item is the assay
if not output[2].lower() in ASSAY_CODES.values():
#is the second item is the assay?
if not output[1].lower() in ASSAY_CODES.values():
print "not a good path", output
project=output[1]+output[3].strip('run')
output=[output[0],output[2],output[1],output[3], output[4],project]
else:
project=output[2]+output[3].strip('run')
output=[output[0],output[1],output[2],output[3], output[4],project]
else:
output=output[-3:]
keys=['date','run', 'project']
pathinfo = dict(zip(keys,output))
pathinfo['date']=munge_date(pathinfo['date'])
pathinfo['project']=pathinfo['project'].lower()
#Set Machine
if re.search('[%s][A-Z]\d+' % ''.join(MACHINE_CODES.keys()), pathinfo['run']):
pathinfo['machine']= MACHINE_CODES[pathinfo['run'][0]]
#Set assay
for a in ASSAY_CODES.keys():
if re.search(a, pathinfo['project'], re.IGNORECASE):
pathinfo['assay'] = ASSAY_CODES[a]
elif re.search('MSI', pathinfo['project'], re.IGNORECASE):
pathinfo['assay'] = 'msi-plus'
#Set prep type
if re.search('kapa', pathinfo['project']):
pathinfo['prep_type']='kapa'
elif re.search('hotspot', pathinfo['assay']):
pathinfo['prep_type']='truseq'
else:
pathinfo['prep_type']='sure_select'
return pathinfo
def munge_transcript(data, RefSeqs):
"""
Return HGVS correct transcript annotations
Filtered with a preferred transcript list
NM_006772.1:c.1713G>A
"""
CODING, PROTEIN = [], []
transcripts = data.get('Transcripts')
if transcripts is not None:
# Split incoming trans, strip the trailing )
data1 = multi_split(transcripts.replace('(', ':'), ',(')
#Remove duplicate transcription entries
data = list(set(data1))
for d in data:
codon, prot, protein, coding, txpt = ' ', ' ', ' ', ' ', None
# Split the actual transcript info which is colon separated
x = d.split(':')
#5: ['PRSS1', 'NM_002769', 'exon4', 'c.567T>C', 'p.L189L']
if len(x)==5:
gene, txpt, exon, codon, prot = x
elif len(x)==4:
#4: ['POLE', 'NM_006231', 'exon25', 'c.2865-4T>-']
gene, txpt, exon, codon = x
elif len(x)==3:
#3: ['RAD50', 'NM_005732', 'c.-38G>A']
if 'NM' in x[1]:
gene, txpt, codon = x
#3: ['NM_005590','exon5','c.315-4T>-']
elif 'NM' in x[0]:
txpt, exon, codon = x
elif len(x)==2:
#2: ['NM_001290310', 'c.*513_*514insATC']
txpt, codon = x
elif len(x)==1:
continue
else:
sys.exit("don't know how to parse %s" % d)
pref_trans = RefSeqs.get(txpt)
#Want to return None for all values if not pref_trans
if not pref_trans:
continue
code = pref_trans + ':' + codon
CODING.append(code)
PROTEIN.append(prot)
return ' '.join(CODING), ' '.join(PROTEIN)
def munge_transcript(data, RefSeqs):
"""
Return HGVS correct transcript annotations
Filtered with a preferred transcript list
NM_006772.1:c.1713G>A
"""
CODING, PROTEIN = [], []
transcripts = data.get('Transcripts')
if transcripts is not None:
# Split incoming trans, strip the trailing )
data1 = multi_split(transcripts.replace('(', ':'), ',(')
#Remove duplicate transcription entries
data = list(set(data1))
for d in data:
codon, prot, protein, coding, txpt = ' ', ' ', ' ', ' ', None
# Split the actual transcript info which is colon separated
x = d.split(':')
#5: ['PRSS1', 'NM_002769', 'exon4', 'c.567T>C', 'p.L189L']
if len(x) == 5:
gene, txpt, exon, codon, prot = x
elif len(x) == 4:
#4: ['POLE', 'NM_006231', 'exon25', 'c.2865-4T>-']
gene, txpt, exon, codon = x
elif len(x) == 3:
#3: ['RAD50', 'NM_005732', 'c.-38G>A']
if 'NM' in x[1]:
gene, txpt, codon = x
#3: ['NM_005590','exon5','c.315-4T>-']
elif 'NM' in x[0]:
txpt, exon, codon = x
elif len(x) == 2:
#2: ['NM_001290310', 'c.*513_*514insATC']
txpt, codon = x
elif len(x) == 1:
continue
else:
sys.exit("don't know how to parse %s" % d)
pref_trans = RefSeqs.get(txpt)
#Want to return None for all values if not pref_trans
if not pref_trans:
continue
code = pref_trans + ':' + codon
CODING.append(code)
PROTEIN.append(prot)
return ' '.join(CODING), ' '.join(PROTEIN)
def match(control, run_info):
"""
Make a list and keep count of variants found in both the qc file and the run output for LMG/OPX-240 sample
Matches if chr, start are the same.
"""
matchedlist = []
nonmatch= []
concount=0
for conline in control:
concount=concount+1
for runline in run_info:
if runline[0].startswith('Position'):
continue
else:
runline=multi_split(runline[0], 'chr:-,')
if (conline[2]==runline[0]) and (int(conline[3])==int(runline[1])):
if conline not in matchedlist:
matchedlist.append(conline)
return matchedlist, concount
def munge_path(pth):
"""
Get run, machine, assay and capture from path
"""
output=multi_split(pth, '/_')
if len(output)<4:
raise ValueError('Incorrect path given. Must be in the format of YYYY-MM-DD_Machine_Assay_Run#_version')
pathinfo={}
run=""
for i in output:
i=i.lower()
if i.startswith('20'):
run=i
elif i.startswith('run'):
pathinfo['run']=run+'_'+i
elif i.startswith('hi') or i.startswith('mi'):
pathinfo['machine']=i
elif i.startswith('onco') or i.startswith('colo'):
pathinfo['assay']=i
elif i.startswith('v'):
pathinfo['capture']=i
return pathinfo
def process_files(infiles, tab, filetype):
"""
Rename the analysis files for workbook
"""
for fname in infiles:
(f_path, f_name) = os.path.split(fname)
if re.search(str(filetype), f_name):
(f_short_name, f_extension) = os.path.splitext(f_name)
sheet_name = multi_split(f_short_name, '._')
try:
#48_A03_BROv7_HA0186_NA12878_QC_Analysis
if sheet_name[-2] == 'QC':
sheet_name = '0_QC'
#48_A03_BROv7_HA0186_NA12878_Quality_Analysis
elif sheet_name[-2] == 'Quality':
sheet_name = '1_QC_Metrics'
#NA12878-HP998-HHv2.Coverage_Gene
elif sheet_name[-3] == 'Coverage':
if sheet_name[-2] == 'Gene':
sheet_name = '15_Gene_Coverage'
elif sheet_name[-2] == 'Exon':
sheet_name = '16_Exon_Coverage'
#48_A03_BROv7_HA0186_NA12878_CNV_QC_[Exon/Gene]_Analysis
elif sheet_name[-3] == 'QC':
if sheet_name[-2] == 'Gene':
sheet_name = '2_QC_by_Gene'
elif sheet_name[-2] == 'Exon':
sheet_name = '3_QC_by_Exon'
#48_A03_BROv7_HA0186_NA12878_CNV_[Exon/Gene]_Analysis
elif sheet_name[-3] == 'CNV':
if sheet_name[-2] == 'Gene':
sheet_name = '7_CNV_Gene'
elif sheet_name[-2] == 'Exon':
sheet_name = '8_CNV_Exon'
#48_A03_BROv7_HA0186_NA12878_SV_Analysis
elif sheet_name[-2] == 'SV':
sheet_name = '4_SV_Gridss'
#48_A03_BROv7_HA0186_NA12878_Breakdancer_Analysis
elif sheet_name[-2] == 'Breakdancer':
sheet_name = '5_SV_Breakdancer'
#48_A03_BROv7_HA0186_NA12878_Pindel_Analysis
elif sheet_name[-2] == 'Pindel':
sheet_name = '6_SV_Pindel'
#48_A03_BROv7_HA0186_NA12878_Genotype_Analysis
elif sheet_name[-2] == 'Genotype':
sheet_name = '9_Clinically_Flagged'
#48_A03_BROv7_HA0186_NA12878_MSI_Analysis
elif sheet_name[-2] == 'MSI':
sheet_name = '12_MSI'
#48_A03_BROv7_HA0186_NA12878_Amplicon_Analysis
elif sheet_name[-2] == 'Amplicon':
sheet_name = '13_Amplicons'
#48_A03_BROv7_HA0186_NA12878_PolyHunter_Analysis
elif sheet_name[-2] == 'PolyHunter':
sheet_name = '14_PolyHunter'
#48_A03_BROv7_HA0186_NA12878_SNP_Indel_Analysis
elif sheet_name[-2] == 'SNP':
sheet_name = '10_SNP'
#48_A03_BROv7_HA0186_NA12878_INDEL_Analysis
elif sheet_name[-2] == 'INDEL':
sheet_name = '11_INDEL'
if sheet_name == tab:
return sheet_name, fname
except IndexError:
continue