# Read source data into list
bioMartList = srcFile.readlines()
# remove trailing '\n' from every record
LEN_bML = len(bioMartList)
i = 0
while i < LEN_bML:
bioMartList[i] = bioMartList[i].rstrip('\n')
i = i + 1
# Grouping records by gene name and splitting record fields into lists
groupedList = JamesDefs.groupByField(bioMartList, 0)
# Combine exon records into a single gene line record with start and stop coords for coding region
# TranscriptID field will be removed and fields representing the number of exons encountered and
# the chromosomal coverage will be appended respectivly to the end of each record
oneLineRecordList = combineExons(groupedList, bdryLen)
# Write out oneLineRecordList to outFile
boundaryFile.writelines(oneLineRecordList)
boundaryFile.close()
print 'Tada!'
t1 = time()
# Populate a Dict with Seq objs for Anopheles boundary seqs
# What follows directly is a klugde to get my seqDict vals to have the IUPAC ambiguous alphabet
boundarySeqs = list(SeqIO.parse(open(boundarySeqs, "rU"), "fasta"))
for record in boundarySeqs :
record.seq.alphabet = IUPACAmbiguousDNA
boundarySeqs = SeqIO.to_dict(boundarySeqs, key_function = lambda rec : rec.description.split()[0])
# convert iupac motifs to regexs and creat list of lists with each motif represented as ['IUPAC', 'REGEX']
convertMotifList(motifList)
# group ClusterDefs by ClusterName
clusterDefinitionList = JamesDefs.groupByField(clusterDefinitionList, 0)
# This will become a list of tab delim'd params for the hyperGeo func: 'Motif:ClusterID';'motifCountInAll';'len(all)';'motifCountInCluster';'numOfSeqsInCluster'
hyperGeoParams_4_motifClusterPairs = []
m=0
for motif in motifList:
m+=1
print 'Motif '+str(m)
# Count how many seq in total list have motif in either orientation
motifCountInAll = None
motifCountInAll = countMotifInAll(motif[1], boundarySeqs)
for cluster in clusterDefinitionList:
# open and create handle for outFile
resFile = open(outFile, 'w')
tick = time.clock()
# read file into list
conflictList = conflictFile.readlines()
# remove trailing '\n' from every record
LEN_cL = len(conflictList)
i = 0
while i < LEN_cL:
conflictList[i] = conflictList[i].rstrip('\n')
i = i + 1
# group file by target gene id using groupByField
fjoinOutByGeneIDList = JamesDefs.groupByField(conflictList, 1)
resolverArgs = {
'strandField' : 4,
'lowerBoundProximal' : 10,
'higherBoundProximal' : 11,
'conflictRegionStrt' : 18,
'conflictRegionEnd' : 19,
'whichBoundary':'upStream'
}
resolvedBoundariesList = resolver(fjoinOutByGeneIDList, resolverArgs)
resFile.writelines(resolvedBoundariesList)
tock = time.clock()
#========================= User Defined Variables =========================
# Path to original file
originalFastaDict = open('/Users/biggus/Documents/James/Data/2KB/2kb_Sequence/2kb_Anopheles/2KBupTSS_goodAffyAGAPsFastasOUT.masked.nr.fas', 'rU')
desiredFastaList = '/Users/biggus/Documents/James/Data/ClusterDefs/TC-Clusters.txt'
outDir = '/Users/biggus/Documents/James/Data/ClusterDefs/TC-Fastas/'
#==========================================================================
desiredFastaList = map(lambda line : line.strip(), open(desiredFastaList, 'rU').readlines())
# Parse clusterDefs into list of clusters
listOfClusterDefs = JamesDefs.groupByField(desiredFastaList,0)
# Instantiate the fasta rec lists with BioPython Seq using geneID field of discriptor as key to seq objects
originalFastaDict = SeqIO.to_dict(SeqIO.parse(originalFastaDict, 'fasta'),
key_function = lambda rec : rec.description.split()[0])
for cluster in listOfClusterDefs:
print "Working on Cluster: %s" % (cluster[0][0])
# New dict to catch copied seqObjs
desiredFastaObjList = []
for rec in cluster:
if originalFastaDict.has_key(rec[1]):