def arenosa_snps(pileup_file, o_1_file): Aa_SNPs = Pileup.parse(pileup_file, region_list="Chr1;Chr2;Chr3;Chr4;Chr5") # Only saves SNP positions by default with gzip.open(o_1_file, 'wb') as outfile: for entry in sorted(Aa_SNPs.pileup_dict.items(), key=lambda entry: (entry[0][0], entry[0][1])): (chr, pos), [ref, nuclist] = entry #nuclist_str = "" #for entry in nuclist: # nuclist_str += str(entry) + "\t" #nuclist_str = nuclist_str[:-1] nuclist_str = '\t'.join([entry for entry in nuclist]) outline = '\t'.join(["At" + str(chr),str(pos),str(ref),str(nuclist_str)]) + "\n" outfile.write(bytes(outline,"UTF-8")) return()
def arenosa_snps(pileup_file, o_1_file):
aa_SNPs = Pileup.parse(pileup_file, region_list="Chr1;Chr2;Chr3;Chr4;Chr5"
) # Only saves SNP positions by default
with gzip.open(o_1_file, 'wb') as outfile:
for entry in sorted(aa_SNPs.pileup_dict.items(),
key=lambda entry: (entry[0][0], entry[0][1])):
(chr, pos), [ref, nuclist] = entry
nuclist_str = '\t'.join([entry for entry in nuclist])
outline = '\t'.join(
["At" + str(chr),
str(pos),
str(ref),
str(nuclist_str)]) + "\n"
outfile.write(bytes(outline, "UTF-8"))
return ()
def filter_sam(sam_file, SNP_file, o_2_file):
pattern = r"(\d+)([MSID])"
recomp = re.compile(pattern)
Aa_SNPs = Pileup.parse(SNP_file, simplified=True, only_save_SNPs=False, region_list="AtChr1;AtChr2;AtChr3;AtChr4;AtChr5")
if sam_file[-3:] == ".gz":
infile = gzip.open(sam_file, 'rb')
else:
infile = open(sam_file)
with gzip.open(o_2_file, 'wb') as outfile:
for line in infile:
if sam_file[-3:] == ".gz": line = str(line, encoding='utf8')
line_split = line.split("\t")
if line_split[0][:1] != "@":
chr = line_split[2]
if chr == "*":
outfile.write(bytes(line,"UTF-8"))
continue
skip_read = False
for i in range(11,len(line_split)):
if line_split[i] == "NM:i:0": # Read has a perfect sequence match
outfile.write(bytes(line,"UTF-8"))
skip_read = True
break
if skip_read == True: continue
read_name = line_split[0] # Delete me
s_pos = int(line_split[3])
cigar = line_split[5]
cigar_pieces = recomp.findall(cigar)
nucs = line_split[9]
current_pos = s_pos
for piece in cigar_pieces:
val, operation = piece
val = int(val)
if operation == "M": # A, C, T, or G
for i in range(current_pos, current_pos + val): # Check all nucleotides for match w/ reference
ref, nuclist = Aa_SNPs.PosInfo(chr, i)
if ref != -1: # Ref == -1 when there isn't an A. arenosa SNP at position (chr, i)
cur_nuc = nucs[i - s_pos:i - s_pos + 1]
if cur_nuc != ref: # Read doesn't contain reference nucleotide
total_nucs = Pileup.sum_nucs(nuclist, True)
SNP = {}
SNP["A"] = nuclist[0]
SNP["C"] = nuclist[1]
SNP["G"] = nuclist[2]
SNP["T"] = nuclist[3]
if cur_nuc in SNP and (SNP[cur_nuc] / total_nucs) >= 0.05:
# SNP present in >=5% of A. arenosa reads mapped to TAIR9, so consider it a valid mismap; skip read
print(str(cur_nuc), " (", str(ref), "): ", str(round(SNP[cur_nuc] / total_nucs,2)), "\t", str(chr), ":", str(i), "\t", str(read_name), sep='') # Delete me
skip_read = True
break
else: # Insertion or deletion
ref, nuclist = Aa_SNPs.PosInfo(chr, i)
if ref != -1: # Ref == -1 when there isn't an A. arenosa SNP at position (chr, i)
if operation == "I":
indel_str = "+" + str(val)
else:
indel_str = "-" + str(val)
for i in range(4,len(nuclist)):
if nuclist[i][:2] == indel_str:
total_nucs = Pileup.sum_nucs(nuclist, True)
if ( int(nuclist[i][3+int(val):]) / total_nucs ) >= 0.05:
# Indel present in >=5% of A. arenosa reads mapped to TAIR9, so consider it a valid mismap; skip read
print(str(nuclist[i]), " (", str(ref), "): ", str(round(int(nuclist[i][3+int(val):]) / total_nucs,2)), str(chr), ":", str(i), "\t", str(read_name), sep='') # Delete me
skip_read = True
break
if skip_read == True:
break
current_pos += val
if skip_read == True:
continue
outfile.write(bytes(line,"UTF-8"))
with open(dup_file) as infile:
for line in infile:
line = line.strip()
dup_list.append(line)
gff_genes_dict = GFF.parse_genes(gff_file)
if adj_value > 0:
for gene_name, [chr, start_pos, end_pos] in gff_genes_dict.gene_dict.items():
start_pos = start_pos - adj_value
end_pos = end_pos + adj_value
gff_genes_dict.gene_dict[gene_name] = [chr, start_pos, end_pos]
if use_random == "Y":
# Construct a list of randomly selected genes to replace the list of duplicated genes
#dup_list = sample(gff_genes_dict.gene_dict.keys(),len(dup_list))
dup_list = gff_genes_dict.gene_dict.keys() # Delete me soon
dup_positions = ";".join([chr + ":" + str(s_pos) + "-" + str(e_pos) for gene_name, [chr,s_pos,e_pos] in sorted(gff_genes_dict.gene_dict.items(), key = lambda x: (x[1][0], x[1][1]) ) if gene_name in dup_list ] )
dup_pileup = Pileup.parse(pileup_file, False, dup_positions)
suns_in_gene = Counter()
suns_in_gene_prob = {}
# SUN-finding algorithm that runs when finding SUNs in As. Looks for 50/50 Con/SNP in As
with open(sun_genes_filename[:-4] + "_SUNs.txt", 'w') as sue_sun_file: # Debugging file
fisher_line = "Pos\tRef\tRefValue\tMaxSNP\tMaxSNPValue\tP-value\n"
sue_sun_file.write(fisher_line)
for entry in dup_pileup:
(chr, pos), [ref, nuclist] = entry
if ref not in ["A","C","G","T"]: continue # Skip ambiguous nucleotide positions
gene_name = gff_genes_dict.get_gene_name(chr, pos)
if gene_name in dup_list:
totalreads, consensus_num, maxSNPs = Pileup.fisher_snp_info(ref,nuclist)
if totalreads < 12 or sum([int(v) for v in maxSNPs.values()]) == 0: continue # Either the total usable reads (consensus + maxSNP) is < 12, or no SNPs are present
def filter_sam(sam_file, SNP_file, o_2_file):
pattern = r"(\d+)([MSID])"
recomp = re.compile(pattern)
aa_SNPs = Pileup.parse(SNP_file,
simplified=True,
only_save_SNPs=False,
region_list="AtChr1;AtChr2;AtChr3;AtChr4;AtChr5")
if sam_file[-3:] == ".gz":
infile = gzip.open(sam_file, 'rb')
else:
infile = open(sam_file)
with gzip.open(o_2_file, 'wb') as outfile:
for line in infile:
if sam_file[-3:] == ".gz": line = line.decode('utf-8')
line_split = line.split("\t")
if line_split[0][:1] != "@":
chr = line_split[2]
if chr == "*":
outfile.write(bytes(line, "UTF-8"))
continue
skip_read = False
for i in range(11, len(line_split)):
if line_split[
i] == "NM:i:0": # Read has a perfect sequence match
outfile.write(bytes(line, "UTF-8"))
skip_read = True
break
if skip_read == True: continue
#read_name = line_split[0] # Delete me
s_pos = int(line_split[3])
cigar = line_split[5]
cigar_pieces = recomp.findall(cigar)
nucs = line_split[9]
current_pos = s_pos
for piece in cigar_pieces:
val, operation = piece
val = int(val)
if operation == "M": # A, C, T, or G
for i in range(
current_pos, current_pos + val
): # Check all nucleotides for match w/ reference
ref, nuclist = aa_SNPs.PosInfo(chr, i)
if ref != -1: # Ref == -1 when there isn't an A. arenosa SNP at position (chr, i)
cur_nuc = nucs[i - s_pos:i - s_pos + 1]
if cur_nuc != ref: # Read doesn't contain reference nucleotide
total_nucs = Pileup.sum_nucs(nuclist, True)
SNP = {}
SNP["A"] = nuclist[0]
SNP["C"] = nuclist[1]
SNP["G"] = nuclist[2]
SNP["T"] = nuclist[3]
if cur_nuc in SNP and (SNP[cur_nuc] /
total_nucs) >= 0.05:
# SNP present in >=5% of A. arenosa reads mapped to TAIR9, so consider it a valid mismap; skip read
#print(str(cur_nuc), " (", str(ref), "): ", str(round(SNP[cur_nuc] / total_nucs,2)), "\t", str(chr), ":", str(i), "\t", str(read_name), sep='') # Delete me
skip_read = True
break
else: # Insertion or deletion
ref, nuclist = aa_SNPs.PosInfo(chr, i)
if ref != -1: # Ref == -1 when there isn't an A. arenosa SNP at position (chr, i)
if operation == "I":
indel_str = "+" + str(val)
else:
indel_str = "-" + str(val)
for i in range(4, len(nuclist)):
if nuclist[i][:2] == indel_str:
total_nucs = Pileup.sum_nucs(nuclist, True)
if (int(nuclist[i][3 + int(val):]) /
total_nucs) >= 0.05:
# Indel present in >=5% of A. arenosa reads mapped to TAIR9, so consider it a valid mismap; skip read
#print(str(nuclist[i]), " (", str(ref), "): ", str(round(int(nuclist[i][3+int(val):]) / total_nucs,2)), str(chr), ":", str(i), "\t", str(read_name), sep='') # Delete me
skip_read = True
break
if skip_read == True:
break
current_pos += val
if skip_read == True:
continue
outfile.write(bytes(line, "UTF-8"))
linelist = line.split()
if line[:1] == ">":
if chr != "":
chr_lengths[chr] = chr_len
chr = linelist[0][1:]
chr_len = 0
else:
line = line.split()
chr_len += len(line[0])
chr_lengths[chr] = chr_len
return(chr_lengths)
pileup, ref, win = command_line()
chr_lengths = get_chr_len(ref)
PileupFile = Pileup.parse(pileup,True)
#windows = [i for i in range(1,1501)]
windows = [i for i in range(1600,5001,100)]
for win in windows:
outdir = "Output/HMMCovWin/" + str(win) + "bp-window/"
if not os.path.exists(outdir):
os.makedirs(outdir)
for chr in sorted(chr_lengths):
histogram = Counter()
outlist = []
for i in range(1,chr_lengths[chr]+1,win):
loc = str(chr) + ":" + str(i) + "-" + str(i+win)
cov = PileupFile.Coverage(loc,True)
if cov in histogram: histogram[cov] += 1
if line[:1] == ">":
if chr != "":
chr_lengths[chr] = chr_len
chr = linelist[0][1:]
chr_len = 0
else:
line = line.split()
chr_len += len(line[0])
chr_lengths[chr] = chr_len
return (chr_lengths)
pileup, ref, win = command_line()
chr_lengths = get_chr_len(ref)
PileupFile = Pileup.parse(pileup, True)
#windows = [i for i in range(1,1501)]
windows = [i for i in range(1600, 5001, 100)]
for win in windows:
outdir = "Output/HMMCovWin/" + str(win) + "bp-window/"
if not os.path.exists(outdir):
os.makedirs(outdir)
for chr in sorted(chr_lengths):
histogram = Counter()
outlist = []
for i in range(1, chr_lengths[chr] + 1, win):
loc = str(chr) + ":" + str(i) + "-" + str(i + win)
cov = PileupFile.Coverage(loc, True)
if cov in histogram: histogram[cov] += 1