if args.reference:
sys.stderr.write("Parsing reference. This could take a while...\n")
try:
with open(args.reference + ".fai", "r") as fai:
scafs_lengths = [line.split()[:2] for line in fai]
except:
sys.stderr.write(
"WARNING: Could not parse fai file, vcf header will not contain contig entries...\n"
)
scafs_lengths = None
with gzip.open(args.reference,
"r") if args.reference.endswith(".gz") else open(
args.reference, "r") as ref:
refDict = dict(zip(*genomics.parseFasta(ref.read())))
else:
refDict = None
#########################################################################################
genoFileReader = genomics.GenoFileReader(genoFile)
allNames = genoFileReader.names
if not args.samples: namesToUse = allNames
else: namesToUse = args.samples.split(",")
outFile.write("##fileformat=VCFv4.2\n")
if refDict:
#args = parser.parse_args("-n 5 -t test.trees -o test.topos.txt -w test.weights.B.csv -g A a,b,c -g B d,e,f -g C g,h,i -g D j,k,l".split())
if args.phylipIn: inFormat = "phylip"
else: inFormat = "fasta"
if args.phylipOut: outFormat = "phylip"
else: outFormat = "fasta"
l = args.lineLen
########################################################################
allText = sys.stdin.read()
if inFormat == "fasta": names, seqs = genomics.parseFasta(allText)
else: names, seqs = genomics.parsePhylip(allText)
if args.truncateNames: names = [name.split()[0] for name in names]
regions = [parseRegionText(r) for r in args.regions] if args.regions else []
if args.regionsFile:
with open(args.regionsFile, "r") as rf:
for line in rf: regions.append(parseRegionList(line.split()))
#only filter and chop sequences if necessary
if len(regions) >= 1:
outNames = []
outSeqs = []
for seqName,start,end,ori in regions:
if args.genoFile:
if args.genoFile[-3:] == ".gz": genoFile = gzip.open(args.genoFile, "w")
else: genoFile = open(args.genoFile, "w")
else: genoFile = sys.stdout
if args.randomPhase:
import random
#############################
#read sequence file
seqString = seqFile.read()
#parse
if args.format == "fasta":
seqNames, seqs = genomics.parseFasta(seqString)
multi = False
elif args.format == "phylip":
#with phylip its possible to have multiple alignments, so we need to check if thats the case
pieces = genomics.parsePhylip(seqString)
if type(pieces) == tuple:
seqNames, seqs = pieces
multi = False
else:
_seqNames_, _seqs_ = zip(*pieces)
multi = True
if not multi:
#if there is a single set of sequences we parse it and output either as contigs or individuals
#sequences to keep
args = parser.parse_args()
################################################################################
#get gene data
sys.stderr.write("Parsing annotation\n")
with gzip.open(args.annotation,
"rt") if args.annotation.endswith(".gz") else open(
args.annotation, "rt") as ann:
geneData = genomics.parseGenes(ann.readlines(), fmt=args.format)
#get scaffold names
sys.stderr.write("Loading reference genome\n")
with gzip.open(args.ref, "rt") if args.ref.endswith(".gz") else open(
args.ref, "rt") as ref:
scaffolds, _sequences_ = genomics.parseFasta(ref.read(),
makeUppercase=True)
sequences = {}
for i, scaffold in enumerate(scaffolds):
sequences[scaffold] = _sequences_[i]
#open output
if not args.outFile: outFile = sys.stdout
else:
outFile = gzip.open(args.outFile,
"wt") if args.outFile.endswith(".gz") else open(
args.outFile, "wt")
outFile.write("\t".join([
"scaffold", "position", "codon_position", "substitution_type", "degeneracy"
]) + "\n")