def main():
''' main scripts '''
bed2anno = defaultdict(list)
bedid2bed = {}
for line in open(sys.argv[1], "r"):
line = line.strip().split("\t")
bed = Bed(line[0:6])
if bed.id not in bedid2bed:
bedid2bed[bed.id] = line[0:6]
gene = genebed(line[6:18])
s = bed_in_bedlist(bed, gene.Exons())
if s == "":
s = bed_in_bedlist(bed, gene.Introns())
if s == "":
sys.exit("### intersection is wrong")
bed2anno[bed.id].append(s)
for k, v in bed2anno.items():
x = "\t".join(bedid2bed[k])
ss = []
ss += ["E" for e in v if "Exon" in e]
ss += ["I" for e in v if "Intron" in e]
if "E" in set(ss) and "I" in set(ss):
alternative = "Alternative_ExonIntron"
elif "E" in set(ss) and "I" not in set(ss):
alternative = "Constitutive_Exon"
elif "I" in set(ss) and "E" not in set(ss):
alternative = "Constitutive_Intron"
else:
pass
print("{0}\t{1}\t{2}\t{3}".format(x, "|".join(v), "".join(ss),
alternative))
def gene_utr_cds_length(genebed_file):
UTR5 = []
ORF = []
UTR3 = []
ncRNA = []
for line in open(genebed_file, "r"):
if not line.startswith("#"):
g = genebed(line.strip().split("\t"))
utr5, cds, utr3 = ORF_UTR_size_of(g)
if cds < 10:
ncRNA.append(utr5 + cds + utr3)
elif cds > 10:
UTR5.append(utr5)
ORF.append(cds)
UTR3.append(utr3)
else:
pass
mean_utr5 = float(np.where(len(UTR5) == 0, 0, np.mean(UTR5)))
mean_utr3 = float(np.where(len(UTR3) == 0, 0, np.mean(UTR3)))
mean_cds = float(np.where(len(ORF) == 0, 0, np.mean(ORF)))
mean_ncRNA = float(np.where(len(ncRNA) == 0, 0, np.mean(ncRNA)))
print("##########################")
print("Average length for features: UTR5, CDS, UTR3, and ncRNA")
print(mean_utr5, mean_cds, mean_utr3, mean_ncRNA)
return mean_utr5, mean_cds, mean_utr3, mean_ncRNA
def Cluster2pIntronRetention(Cluster):
''''clu_5077': {'chrX\t166458208\t166462315': 0.0195360774825754, 'chrX\t166458343\t166462315': -0.0195360774825754}
'''
for k1, v1 in Cluster.items():
if len(v1) > 2:
for k2, v2 in v1.items():
Junction = BedTool(k2, from_string=True)
Junction_Bed = Bed(k2.split("\t"))
X = Junction.intersect(Genes, wb=True)
m6A_string = []
splicing_type = []
if len(X) >= 1:
for gene in X:
g = genebed(gene[3:])
for e in g.Exons():
if Junction_Bed.overlap(Junction_Bed, e):
x_l = Junction_Bed.overlapLength(e)
if x_l >= 10 and x_l < e.length(
) - 2: ## if ==, which means the entire exon is inside of the intron
exon = BedTool(str(e), from_string=True)
alternative = Junction.intersect(exon)
splicing_type.append("pfIntronRetention")
m6A_string.append(m6AORnot(alternative))
if len(set(m6A_string)) == 1 and len(splicing_type) > 0:
fo_pfIR.write("{0}\t{1}\t{2}\t{3}\t{4}\n".format(
"|".join(set(splicing_type)),
"|".join(set(m6A_string)), k1, k2, v2))
def main():
''' main scripts '''
summit2bed = summitTobed("../m6AIP_Cfg1_Cfg2.narrowPeak.refined.bed")
for line in open(sys.argv[1], 'r'):
line = line.strip().split("\t")
### from the bed.id to bed interval
try:
bed = summit2bed[line[3]]
except:
sys.exit("# Something is wrong with the summit2bed")
gene = genebed(line[6:18])
Introns = gene.Introns()
random_n = random.randint(0, len(Introns) - 1)
f1 = bed_in_Intron(Bed(line[0:6]), [Bed(line[6:12])])
random_start = random.randint(gene.start, gene.end)
f2 = bed_in_Intron(
Bed([
bed.chr, random_start, random_start + 1, "None", 0, bed.strand
]), [Bed(line[6:12])])
print("{0}\t{1}".format(f1, f2))
def main():
''' main scripts '''
summit2bed = summitTobed("../m6AIP_Cfg1_Cfg2.narrowPeak.refined.bed")
for line in open(sys.argv[1], 'r'):
line = line.strip().split("\t")
### from the bed.id to bed interval
try:
bed = summit2bed[line[3]]
except:
sys.exit("# Something is wrong with the summit2bed")
gene = genebed(line[6:18])
Introns = gene.Introns()
random_n = random.randint(0, len(Introns) - 1)
for i, intron in enumerate(Introns):
if bed.overlap(bed, intron):
L = bed.overlapLength(intron)
f = bed_in_Intron(Bed(line[0:6]), [intron])
overlap_start, overlap_end = max(bed.start, intron.start), min(
bed.end, intron.end)
try:
random_start = random.randint(
0,
intron.length() - 1 - L) + intron.start
random_end = random_start + L
except:
while i > 0:
X = Introns[i - 1]
if X.length() >= L + 1:
random_start = random.randint(
0,
X.length() - 1 - L) + X.start
random_end = random_start + L
break
else:
i = i - 1
if f + 1 > 0:
print("{0}\t{1}\t{2}\t{3}\t{4}\t{5}\t{6}\t{7}\t{8}".format(
bed, intron, f,
overlap_start, overlap_end, random_start, random_end,
intron.length(), Introns[random_n].length()))
def utr5_cds_utr3_rPos(intersect_file):
''' if the overlapped genebed is protein_coding gene and return the relative position to the feature start '''
''' for example: UTR5 is [0,1220], and the single-point is 125, so it returns 125/1220=0.10245902'''
UTR5_rpos = []
CDS_rpos = []
UTR3_rpos = []
ncRNA_rpos = []
Intron_rpos = []
for line in open(intersect_file, "r"):
if not line.startswith("#"):
line = line.strip().split("\t")
b = Bed(line[0:6])
g = genebed(line[6:18])
## print("\t".join(line)) ## this is test line
utr5, cds, utr3 = ORF_UTR_size_of(g)
## print(utr5, cds, utr3) ## this is test line
#### Introns
if g.Introns():
f = bed_in_Intron(b, g.Introns())
## print("{0}\t{1}".format("test ", f)) ## this si test line
if f >= 0:
Intron_rpos.append(f)
if cds < 10:
fo.write("{0}\t{1}\t{2}\n".format(
"\t".join(line), "ncRNA", "intron"))
else:
fo.write("{0}\t{1}\t{2}\n".format(
"\t".join(line), "protein_coding", "intron"))
continue
########
if cds < 10:
ncRNA_rpos.append(bed_in_bedlist(b, g.Exons()))
fo.write("{0}\t{1}\t{2}\n".format("\t".join(line), "ncRNA",
"exon"))
else:
try:
if g.CDS().start <= b.start < b.end <= g.CDS().end:
CDS_rpos.append(
bed_in_bedlist(bed=b, bedlist=g.CDS().Exons()))
elif utr5 > 0 and g.UTR5(
).start <= b.start < b.end <= g.UTR5().end:
UTR5_rpos.append(
bed_in_bedlist(bed=b, bedlist=g.UTR5().Exons()))
elif utr3 > 0 and g.UTR3(
).start <= b.start < b.end <= g.UTR3().end:
UTR3_rpos.append(
bed_in_bedlist(bed=b, bedlist=g.UTR3().Exons()))
else:
pass
fo.write("{0}\t{1}\t{2}\n".format("\t".join(line),
"protein_coding",
"exon"))
except AttributeError:
pass
print("####################")
print("# Intron peaks:")
print(len(Intron_rpos))
print("# ncRNA peaks: ")
print(len(ncRNA_rpos))
print("# 5UTR peaks:")
print(len(UTR5_rpos))
print("# CDS peaks:")
print(len(CDS_rpos))
print("# 3UTR peaks:")
print(len(UTR3_rpos))
return UTR5_rpos, CDS_rpos, UTR3_rpos, ncRNA_rpos, Intron_rpos
def main():
''' '''
size_of_ORF = {}
size_of_RNA = {}
transcript_bed = {}
args = parse_argument()
mode = args.mode
genefile = args.genebed
if mode == "m":
fo = open(
genefile.strip(".GeneBed") + ".MaxORF_LongestNcRNA.GeneBed", "w")
print("## Searching the maxORF and Longest Transcript ... ",
file=sys.stderr)
elif mode == "r":
fo = open(
genefile.strip(".GeneBed") + ".randomTranscript.GeneBed", "w")
print("## Searching the random Transcript ... ", file=sys.stderr)
else:
print("## Error in the isoform selection mode ", file=sys.stderr)
print("## Calculating the size of ORF and gene length ... ",
file=sys.stderr)
## loading the GeneBed file
for x in Iter_filehandle(filename=genefile):
gene = genebed(x)
## gene.id in genebed class is actually transcript_id
transcript_bed[gene.id] = x
#### for the gene without "NM_" or "NR_" in name
size_of_ORF[gene.id] = ORF_UTR_size_of_v2(gene)[1]
size_of_RNA[gene.id] = length_of_transcript(gene)
##
### geneid_genename_transcriptid_genetype.txt file
for x in Iter_filehandle(
filename="geneid_genename_transcriptid_genetype.txt"):
#### Geneid, GeneName, Transcripts, GeneType
genename = x[1]
transcripts = x[2].split('|')
longest_one = ''
orf = []
#### gene_type
try:
if x[3]:
pass
except IndexError:
x.append("")
##
for t in transcripts:
try:
orf.append(size_of_ORF[t])
except:
print("## Error for calculating the ORF size for, ",
t,
file=sys.stderr)
## max ORF, if equal orf, long transcript
m = max(orf)
index = []
for idx, size in enumerate(orf):
if size == m:
index.append(idx)
if len(index) == 1:
longest_one = transcripts[orf.index(m)]
else:
RNALength = []
for i, t in enumerate(transcripts):
if i in index:
RNALength.append(size_of_RNA[t])
longest_one = transcripts[index[RNALength.index(max(RNALength))]]
random_index = random.randint(0, len(transcripts) - 1)
random_one = transcripts[random_index]
if mode == "m":
select_one = longest_one
elif mode == "r":
select_one = random_one
else:
print("## Error in the isoform selection mode ", file=sys.stderr)
select_genebed = transcript_bed[select_one]
if x[0] == x[1] and not x[3]:
fo.write("{0}\t{1}\t{2}\n".format("\t".join(select_genebed[0:3]),
select_one + "|" + genename,
"\t".join(select_genebed[4:])))
elif x[0] == x[1] and x[3]:
fo.write("{0}\t{1}\t{2}\n".format(
"\t".join(select_genebed[0:3]),
select_one + "|" + genename + "|" + x[3],
"\t".join(select_genebed[4:])))
elif x[0] != x[1] and not x[3]:
fo.write("{0}\t{1}\t{2}\n".format(
"\t".join(select_genebed[0:3]),
select_one + "|" + genename + "|" + x[0],
"\t".join(select_genebed[4:])))
else:
fo.write("{0}\t{1}\t{2}\n".format(
"\t".join(select_genebed[0:3]),
select_one + "|" + genename + "|" + x[0] + "|" + x[3],
"\t".join(select_genebed[4:])))
print("## Done ", file=sys.stderr)
fo.close()