def bridgemapper(istm, ostm, strMap, strCols, strFrom, strTo, ostmLog, iSkip):
strCols = strCols[1:-1]
aiCols = [int(s) for s in re.split(r'\s*,\s*', strCols)] if strCols else []
#Open a blank metadata object and initialize
pMeta = metadata.open()
pMeta.set("mapped", False)
aastrData = []
setstrIn = set()
csvr = csv.reader(istm, csv.excel_tab)
for astrLine in csvr:
aastrData.append(astrLine)
pMeta.update_md5sum("\t".join(astrLine))
if csvr.line_num < iSkip:
continue
for iCol in aiCols:
if iCol < len(astrLine):
setstrIn.add(astrLine[iCol])
else:
sys.stderr.write(
" +++ ERROR in GeneMapper +++ Number of requested columns to \
map is larger than the number of columns in the input data file.\n")
pMeta.store_checksum()
hashMap = None
# Make sure mapping file exists and has nonzero file size
if strMap and os.path.exists(strMap) and (os.stat(strMap)[6] > 0):
hashMap = convertGeneIds(setstrIn, strMap, strFrom, strTo)
else:
sys.stderr.write(
"+++ ERROR in GeneMapper +++ Input file does not exist or is empty. \
Return empty file. \n")
if hashMap:
if any(hashMap.values()):
pMeta.set("mapped", True)
for iRow in range(iSkip, len(aastrData)):
astrRow = aastrData[iRow]
for iCol in aiCols:
if iCol < len(astrRow):
strTo = hashMap.get(astrRow[iCol])
if strTo:
astrRow[iCol] = strTo
else:
astrRow[iCol] = ""
else:
sys.stderr.write("+++Error in GeneMapper +++ Empty mapping. \
Return original file. \n")
csvw = csv.writer(ostm, csv.excel_tab)
#make sure that if the mapping is empty for one of the columns, delete the entire row
for astrLine in aastrData:
if all(astrLine): csvw.writerow(astrLine)
if ostmLog:
pMeta.save_text(ostmLog)
def _main():
args = argp.parse_args()
iLC = sfle.lc(args.istm.name)
if (args.strFrom == args.strTo) or (iLC >= args.iMaxLines):
#if the two gene identifier types are the same, or if the line count exceeds iMaxLines,
#return the input file
pAastrData = csv.reader(args.istm, csv.excel_tab)
csvw = csv.writer(args.ostm, csv.excel_tab)
pMeta = metadata.open()
for astrLine in pAastrData:
csvw.writerow(astrLine)
pMeta.update_md5sum("\t".join(astrLine))
pMeta.store_checksum()
if args.ostmLog:
pMeta = metadata.open()
pMeta.set("mapped", True)
pMeta.save_text(args.ostmLog)
elif not (args.strMap) or not (os.stat(str(args.strMap))[6]):
#if there is no map file specified
pAastrData = csv.reader(args.istm, csv.excel_tab)
csvw = csv.writer(args.ostm, csv.excel_tab)
pMeta = metadata.open()
for astrLine in pAastrData:
csvw.writerow(astrLine)
pMeta.update_md5sum("\t".join(astrLine))
pMeta.store_checksum()
if args.ostmLog:
pMeta.set("mapped", False)
pMeta.save_text(args.ostmLog)
else:
#if gene sniffer flag is on, try to guess the best possible gene identifier
if args.fSniffer:
args.strFrom = gene_sniffer(args.istm, args.strCols)
bridgemapper(args.istm, args.ostm, args.strMap, args.strCols,
args.strFrom, args.strTo, args.ostmLog, args.iSkip)
def makeunique(istm, ostm, strSplit, iCols, iSkip, ostmLog):
'''
Splits up compressed elements (e.g. a///b c///d) into their cartesian products (e.g. a c, a d, b c, bd);
then, gets rid of duplicate elements in the set of unordered tuples (equivalence under permutation).
User can specify how many lines to skip beforehand, and how many columns are to be considered
'''
aastrMatIn = [x for x in csv.reader(istm, csv.excel_tab)]
aastrHeaders, aastrDataIn = aastrMatIn[:iSkip], aastrMatIn[iSkip:]
#open a blank metadata object
pMeta = metadata.open()
if strSplit:
aastrSplit = []
for astrRow in aastrDataIn:
astrNames, astrVals = astrRow[:iCols], astrRow[iCols:]
#Choice: keep rows without values
if any(astrNames) and reduce(
lambda y, z: y or z,
[x.find(strSplit) != -1 for x in astrNames]):
astrSplit = [[y.strip() for y in x.split(strSplit)]
for x in astrNames]
aastrSplit += [
list(v) + astrVals
for v in [x for x in itertools.product(*astrSplit)]
]
else:
aastrSplit.append(astrRow)
else:
aastrSplit = aastrDataIn
pNames = set([])
aastrUnique = []
for astrRow in aastrSplit:
astrNames, astrVals = astrRow[:iCols], astrRow[iCols:]
lenNames = len(astrNames)
if (len(frozenset(astrNames)) == lenNames
and not (frozenset(astrNames) in pNames)):
pNames |= {frozenset(astrNames)}
aastrUnique.append(astrRow)
pMeta.set("mapped", any(aastrUnique))
#write output
csvw = csv.writer(ostm, csv.excel_tab)
for row in aastrHeaders + aastrUnique:
csvw.writerow(row)
#save metadata
if ostmLog:
pMeta.save_text(ostmLog)
import arepa
import csv
import sys
import metadata
c_strCurated = "curated"
if not (2 <= len(sys.argv[1:]) <= 3):
raise Exception(
"usage: pkl2metadata.py <ID.pkl> <per-exp.txt> [per-cond.txt]")
c_fileIDpkl, c_fileExpTable = sys.argv[1:3]
c_fileCondTable = sys.argv[3] if len(sys.argv[1:]) > 2 else None
hashMeta = metadata.open(open(c_fileIDpkl, "r"))
def writeTable(hMeta, astrKeys, outputf, bIter=False):
hMeta = {k: hMeta[k] for k in astrKeys}
csvw = csv.writer(open(outputf, "w"), csv.excel_tab)
if bIter:
astrHeader = hMeta.get("sample_name") or hMeta.get("")
_astrKeys = [x for x in astrKeys if hMeta.get(x) != astrHeader]
csvw.writerow(["sample_name"] + _astrKeys)
for iSample, strSample in enumerate(astrHeader):
csvw.writerow( [str(x).replace("\n", " ") for x in [ strSample ] + [hMeta.get(s)[iSample] for s in \
_astrKeys]])
else:
csvw.writerow(astrKeys)
csvw.writerow(
strMethods, strAuthors, strPMIDs, strTaxAs, strTaxBs, strTypes,
strDBs, strIDs, strConfs):
metadatum(pMetadata.taxid, [strTaxAs, strTaxBs], 1)
metadatum(pMetadata.pmid, [strPMIDs], 1)
metadatum(pMetadata.type, [strTypes.lower()], 2)
metadatum(pMetadata.technique, [strMethods], 2)
if len(sys.argv) < 2:
raise Exception("Usage: c2metadata.py <id> < <mpidbc>")
strTarget = sys.argv[1]
strStatus = sys.argv[2] if len(sys.argv[1:]) > 1 else None
strManual = sys.argv[3] if len(sys.argv[1:]) > 2 else None
pMetadata = metadata.open()
cfile.read(sys.stdin, c_iColumns, strTarget, callback, pMetadata)
if strStatus:
strMapped, strBool = [
x for x in csv.reader(open(strStatus), csv.excel_tab)
][0]
fMapped = True if strBool == "True" else False
pMetadata.set(strMapped, fMapped)
if strManual:
for strKey, strVal in csv.reader(open(strManual), csv.excel_tab):
pMetadata.set(strKey, strVal)
pMetadata.save(sys.stdout)