def count_alignments(bam_filename):
import subprocess
from Betsy import module_utils as mlib
samtools = mlib.findbin("samtools")
x = [samtools, "view", bam_filename]
p = subprocess.Popen(x,
bufsize=0,
stdin=subprocess.PIPE,
stdout=subprocess.PIPE,
stderr=subprocess.STDOUT,
close_fds=True)
r = p.stdout
alignments = 0
aligned_reads = {}
for line in r:
# M03807:17:000000000-AHGYH:1:2108:11122:14861 99 1 14172 0 12S128...
# ST-J00106:110:H5NY5BBXX:4:1101:1702:1209 2:N:0:NTCACG 141 *
x = line.split("\t")
assert len(x) >= 11, "SAM format"
qname, flag = x[:2]
flag = int(flag)
# 2 mapped in proper pair
# 4 query is unmapped
# 8 mate is unmapped
#is_aligned = flag & 0x02
is_aligned = not (flag & 0x04)
if is_aligned:
alignments += 1
aligned_reads[qname] = 1
aligned_reads = len(aligned_reads)
return aligned_reads, alignments
def run(self, network, antecedents, out_attributes, user_options,
num_cores, out_path):
import os
from genomicode import parallel
from genomicode import alignlib
from genomicode import filelib
from Betsy import module_utils as mlib
bam_node, ref_node = antecedents
bam_filenames = mlib.find_bam_files(bam_node.identifier)
assert bam_filenames, "No .bam files."
ref = alignlib.create_reference_genome(ref_node.identifier)
filelib.safe_mkdir(out_path)
metadata = {}
metadata["tool"] = "samtools %s" % alignlib.get_samtools_version()
# list of (in_filename, err_filename, out_filename)
jobs = []
for in_filename in bam_filenames:
p, f = os.path.split(in_filename)
sample, ext = os.path.splitext(f)
err_filename = os.path.join(out_path, "%s.log" % sample)
out_filename = os.path.join(out_path, "%s.pileup" % sample)
x = in_filename, err_filename, out_filename
jobs.append(x)
# samtools mpileup -f [reference sequence] [BAM file(s)]
# > myData.mpileup
samtools = mlib.findbin("samtools")
sq = mlib.sq
commands = []
for x in jobs:
in_filename, err_filename, out_filename = x
x = [
sq(samtools),
"mpileup",
"-f",
sq(ref.fasta_file_full),
]
x.append(sq(in_filename))
x = " ".join(map(str, x))
x = "%s 2> %s 1> %s" % (x, err_filename, out_filename)
commands.append(x)
parallel.pshell(commands, max_procs=num_cores)
metadata["num_cores"] = num_cores
metadata["commands"] = commands
x = [x[-1] for x in jobs]
filelib.assert_exists_nz_many(x)
return metadata
def run(self, network, in_data, out_attributes, user_options, num_cores,
out_path):
import os
from genomicode import filelib
from genomicode import parallel
from Betsy import module_utils as mlib
sam_filenames = mlib.find_sam_files(in_data.identifier)
assert sam_filenames, "No .sam files."
filelib.safe_mkdir(out_path)
metadata = {}
samtools = mlib.findbin("samtools")
jobs = [] # list of (sam_filename, bam_filename)
for sam_filename in sam_filenames:
p, f = os.path.split(sam_filename)
assert f.endswith(".sam")
f = f.replace(".sam", ".bam")
bam_filename = os.path.join(out_path, f)
x = sam_filename, bam_filename
jobs.append(x)
# Make a list of samtools commands.
sq = parallel.quote
commands = []
for x in jobs:
sam_filename, bam_filename = x
# samtools view -bS -o <bam_filename> <sam_filename>
x = [
sq(samtools),
"view",
"-bS",
"-o",
sq(bam_filename),
sq(sam_filename),
]
x = " ".join(x)
commands.append(x)
metadata["commands"] = commands
metadata["num_cores"] = num_cores
parallel.pshell(commands, max_procs=num_cores)
# Make sure the analysis completed successfully.
x = [x[-1] for x in jobs]
filelib.assert_exists_nz_many(x)
return metadata
def run(self, network, in_data, out_attributes, user_options, num_cores,
out_path):
import os
from genomicode import filelib
from genomicode import parallel
from Betsy import module_utils as mlib
filenames = mlib.find_fastq_files(in_data.identifier)
assert filenames, "FASTQ files not found: %s" % in_data.identifier
filelib.safe_mkdir(out_path)
metadata = {}
fastqc = mlib.findbin("fastqc")
fastqc_q = parallel.quote(fastqc)
commands = [
"%s --outdir=%s --extract %s" % (fastqc_q, out_path, x)
for x in filenames
]
metadata["commands"] = commands
metadata["num_cores"] = num_cores
#commands = ["ls > %s" % x for x in filenames]
parallel.pshell(commands, max_procs=num_cores)
# Fastqc generates files:
# <file>_fastqc/
# <file>_fastqc.zip
# The contents of the .zip file are identical to the directories.
# If this happens, then delete the .zip files because they are
# redundant.
files = os.listdir(out_path)
filenames = [os.path.join(out_path, x) for x in files]
for filename in filenames:
zip_filename = "%s.zip" % filename
if os.path.exists(zip_filename):
os.unlink(zip_filename)
def run(
self, network, in_data, out_attributes, user_options, num_cores,
out_path):
import os
import shutil
from genomicode import parallel
from genomicode import filelib
from genomicode import alignlib
from Betsy import module_utils as mlib
bam_filenames = mlib.find_bam_files(in_data.identifier)
filelib.safe_mkdir(out_path)
metadata = {}
metadata["tool"] = "bam2fastx (unknown version)"
# Somehow bam2fastx doesn't work if there are spaces in the
# filename. Make a temporary filename with no spaces, and
# then rename it later.
# Actually, may not be bam2fastx's fault.
jobs = []
for i, bam_filename in enumerate(bam_filenames):
p, f, e = mlib.splitpath(bam_filename)
#bai_filename = alignlib.find_bai_file(bam_filename)
#assert bai_filename, "Missing index for: %s" % bam_filename
#temp_bam_filename = "%d.bam" % i
#temp_bai_filename = "%d.bam.bai" % i
#temp_fa_filename = "%d.fa" % i
fa_filename = os.path.join(out_path, "%s.fa" % f)
x = filelib.GenericObject(
bam_filename=bam_filename,
#bai_filename=bai_filename,
#temp_bam_filename=temp_bam_filename,
#temp_bai_filename=temp_bai_filename,
#temp_fa_filename=temp_fa_filename,
fa_filename=fa_filename)
jobs.append(x)
bam2fastx = mlib.findbin("bam2fastx")
# Link all the bam files.
#for j in jobs:
# assert not os.path.exists(j.temp_bam_filename)
# #assert not os.path.exists(j.temp_bai_filename)
# os.symlink(j.bam_filename, j.temp_bam_filename)
# #os.symlink(j.bai_filename, j.temp_bai_filename)
commands = []
for j in jobs:
# bam2fastx -A --fasta -o rqc14.fa rqc11.bam
x = [
mlib.sq(bam2fastx),
"-A",
"--fasta",
#"-o", mlib.sq(j.temp_fa_filename),
#mlib.sq(j.temp_bam_filename),
"-o", mlib.sq(j.fa_filename),
mlib.sq(j.bam_filename),
]
x = " ".join(x)
commands.append(x)
metadata["commands"] = commands
metadata["num_cores"] = num_cores
parallel.pshell(commands, max_procs=num_cores)
#for j in jobs:
# # Move the temporary files to the final location.
# shutil.move(j.temp_fa_filename, j.fa_filename)
# # Remove the link to the BAM file.
# os.unlink(j.temp_bam_filename)
x = [j.fa_filename for x in jobs]
filelib.assert_exists_nz_many(x)
return metadata
def run(self, network, antecedents, out_attributes, user_options,
num_cores, outfile):
from genomicode import parselib
from genomicode import parallel
from Betsy import module_utils as mlib
MAX_CORES = 4 # I/O intensive.
fastq_node, sample_node, bam_node = antecedents
bam_filenames = mlib.find_bam_files(bam_node.identifier)
sample2fastq = mlib.find_merged_fastq_files(sample_node.identifier,
fastq_node.identifier,
as_dict=True)
metadata = {}
jobs = [] # list of (sample, bam_file, fastq_file)
for filename in bam_filenames:
path, sample, ext = mlib.splitpath(filename)
assert sample in sample2fastq, "Missing fastq: %s" % sample
fastq1, fastq2 = sample2fastq[sample]
x = sample, filename, fastq1
jobs.append(x)
funcalls = []
for x in jobs:
sample, bam_filename, fastq_filename = x
# Count the number of reads.
x1 = count_reads, (fastq_filename, ), {}
# Count the number of alignments.
x2 = count_alignments, (bam_filename, ), {}
funcalls.append(x1)
funcalls.append(x2)
assert len(funcalls) == len(jobs) * 2
nc = min(num_cores, MAX_CORES)
results = parallel.pyfun(funcalls, num_procs=nc)
metadata["num_cores"] = nc
# list of (sample, aligns, aligned_reads, total_reads, perc_aligned).
results2 = []
for i, x in enumerate(jobs):
sample, bam_filename, fastq_filename = x
x1 = results[i * 2]
x2 = results[i * 2 + 1]
total_reads = x1
aligned_reads, alignments = x2
perc_aligned = float(aligned_reads) / total_reads
x = sample, alignments, aligned_reads, total_reads, perc_aligned
results2.append(x)
results = results2
# sort by sample name
results.sort()
# Make table where the rows are the samples and the columns
# are the statistics.
table = []
header = ("Sample", "Alignments", "Aligned Reads", "Total Reads",
"Perc Aligned")
table.append(header)
for x in results:
sample, alignments, aligned_reads, total_reads, perc_aligned = x
x1 = parselib.pretty_int(alignments)
x2 = parselib.pretty_int(aligned_reads)
x3 = parselib.pretty_int(total_reads)
x4 = "%.2f%%" % (perc_aligned * 100)
x = sample, x1, x2, x3, x4
assert len(x) == len(header)
table.append(x)
# Write out the table as text file.
TXT_FILE = "summary.txt"
handle = open(TXT_FILE, 'w')
for x in table:
print >> handle, "\t".join(x)
handle.close()
txt2xls = mlib.findbin("txt2xls", quote=True)
parallel.sshell("%s -b %s > %s" % (txt2xls, TXT_FILE, outfile))
return metadata
def run(self, network, antecedents, out_attributes, user_options,
num_cores, out_path):
import os
from genomicode import filelib
from genomicode import parallel
from genomicode import alignlib
from Betsy import module_utils as mlib
bam_node, nc_node, ref_node = antecedents
bam_filenames = mlib.find_bam_files(bam_node.identifier)
assert bam_filenames, "No .bam files."
nc_match = mlib.read_normal_cancer_file(nc_node.identifier)
ref = alignlib.create_reference_genome(ref_node.identifier)
filelib.safe_mkdir(out_path)
metadata = {}
metadata["tool"] = "MuSE %s" % alignlib.get_muse_version()
wgs_or_wes = mlib.get_user_option(user_options,
"wgs_or_wes",
not_empty=True,
allowed_values=["wgs", "wes"])
dbsnp_file = mlib.get_user_option(user_options,
"muse_dbsnp_vcf",
not_empty=True,
check_file=True)
# Make sure dbsnp_file is compressed and indexed.
assert dbsnp_file.endswith(".vcf.gz"), \
"muse_dbsnp_vcf must be bgzip compressed."
x = "%s.tbi" % dbsnp_file
assert filelib.exists_nz(x), "muse_dbsnp_vcf must be tabix indexed."
# sample -> bam filename
sample2bamfile = mlib.root2filename(bam_filenames)
# Make sure files exist for all the samples.
mlib.assert_normal_cancer_samples(nc_match, sample2bamfile)
# list of (normal_sample, cancer_sample, normal_bamfile, tumor_bamfile,
# muse_call_stem, muse_call_file, raw_vcf_outfile, vcf_outfile,
# logfile1, logfile2)
opj = os.path.join
jobs = []
for (normal_sample, cancer_sample) in nc_match:
normal_bamfile = sample2bamfile[normal_sample]
cancer_bamfile = sample2bamfile[cancer_sample]
path, sample, ext = mlib.splitpath(cancer_bamfile)
muse_call_stem = opj(out_path, "%s.call" % cancer_sample)
muse_call_file = "%s.MuSE.txt" % muse_call_stem
raw_vcf_outfile = opj(out_path, "%s.vcf.raw" % cancer_sample)
vcf_outfile = opj(out_path, "%s.vcf" % cancer_sample)
log_outfile1 = opj(out_path, "%s.call.log" % cancer_sample)
log_outfile2 = opj(out_path, "%s.sump.log" % cancer_sample)
x = normal_sample, cancer_sample, normal_bamfile, cancer_bamfile, \
muse_call_stem, muse_call_file, raw_vcf_outfile, vcf_outfile, \
log_outfile1, log_outfile2
jobs.append(x)
# Generate the commands.
# MuSE call -O test11 -f genomes/Broad.hg19/Homo_sapiens_assembly19.fa\
# bam04/196B-MG.bam bam04/PIM001_G.bam
# MuSE sump -I test11.MuSE.txt -E -O test12.vcf \
# -D MuSE/dbsnp_132_b37.leftAligned.vcf.gz
MuSE = mlib.findbin("muse")
sq = mlib.sq
commands = []
for x in jobs:
normal_sample, cancer_sample, normal_bamfile, cancer_bamfile, \
muse_call_stem, muse_call_file, raw_vcf_outfile, vcf_outfile, \
log_outfile1, log_outfile2 = x
x = [
sq(MuSE),
"call",
"-O",
muse_call_stem,
"-f",
sq(ref.fasta_file_full),
cancer_bamfile,
normal_bamfile,
]
x = " ".join(x)
x = "%s >& %s" % (x, log_outfile1)
commands.append(x)
assert len(commands) == len(jobs)
# Not sure about RAM.
nc = mlib.calc_max_procs_from_ram(10, upper_max=num_cores)
parallel.pshell(commands, max_procs=nc)
metadata["num_cores"] = nc
metadata["commands"] = commands
# Make sure the log files have no errors. The files should be
# empty.
log_files = [x[8] for x in jobs]
filelib.assert_exists_z_many(log_files)
# Make sure the call files are created and not empty.
call_files = [x[5] for x in jobs]
filelib.assert_exists_nz_many(call_files)
# Run the "sump" step.
commands = []
for x in jobs:
normal_sample, cancer_sample, normal_bamfile, cancer_bamfile, \
muse_call_stem, muse_call_file, raw_vcf_outfile, vcf_outfile, \
log_outfile1, log_outfile2 = x
x = [
sq(MuSE),
"sump",
"-I",
sq(muse_call_file),
]
assert wgs_or_wes in ["wgs", "wes"]
if wgs_or_wes == "wgs":
x += ["-G"]
else:
x += ["-E"]
x += [
"-O",
sq(raw_vcf_outfile),
"-D",
sq(dbsnp_file),
]
x = " ".join(x)
x = "%s >& %s" % (x, log_outfile2)
commands.append(x)
assert len(commands) == len(jobs)
# Not sure about RAM.
nc = mlib.calc_max_procs_from_ram(10, upper_max=num_cores)
parallel.pshell(commands, max_procs=nc)
metadata["commands"] = metadata["commands"] + commands
# Make sure the log files have no errors. The files should be
# empty.
log_files = [x[9] for x in jobs]
filelib.assert_exists_z_many(log_files)
# Make sure the raw files are created and not empty.
vcf_files = [x[6] for x in jobs]
filelib.assert_exists_nz_many(vcf_files)
# Fix the files.
commands = [] # Should be python commands.
for x in jobs:
normal_sample, cancer_sample, normal_bamfile, cancer_bamfile, \
muse_call_stem, muse_call_file, raw_vcf_outfile, vcf_outfile, \
log_outfile1, log_outfile2 = x
args = normal_sample, cancer_sample, raw_vcf_outfile, vcf_outfile
x = alignlib.clean_muse_vcf, args, {}
commands.append(x)
parallel.pyfun(commands, num_procs=num_cores)
# Delete the log_outfiles if empty.
for x in jobs:
normal_sample, cancer_sample, normal_bamfile, cancer_bamfile, \
muse_call_stem, muse_call_file, raw_vcf_outfile, vcf_outfile, \
log_outfile1, log_outfile2 = x
if os.path.exists(log_outfile1):
os.unlink(log_outfile1)
if os.path.exists(log_outfile2):
os.unlink(log_outfile2)
# Make sure output VCF files exist.
x = [x[7] for x in jobs]
filelib.assert_exists_many(x)
return metadata
def run(self, network, antecedents, out_attributes, user_options,
num_cores, out_path):
import os
from genomicode import parallel
from genomicode import filelib
from genomicode import alignlib
from genomicode import hashlib
from Betsy import module_utils as mlib
fastq_node, sample_node, orient_node, reference_node = antecedents
fastq_files = mlib.find_merged_fastq_files(sample_node.identifier,
fastq_node.identifier)
ref = alignlib.create_reference_genome(reference_node.identifier)
assert os.path.exists(ref.fasta_file_full)
orient = mlib.read_orientation(orient_node.identifier)
filelib.safe_mkdir(out_path)
metadata = {}
metadata["tool"] = "bowtie2 %s" % alignlib.get_bowtie2_version()
# Bowtie2 doesn't handle files with spaces in them. Make
# temporary files without spaces.
# Make a list of the jobs to run.
jobs = []
for i, x in enumerate(fastq_files):
sample, pair1, pair2 = x
bam_filename = os.path.join(out_path, "%s.bam" % sample)
log_filename = os.path.join(out_path, "%s.log" % sample)
sample_h = hashlib.hash_var(sample)
temp_pair1 = "%d_%s_1.fa" % (i, sample_h)
temp_pair2 = None
if pair2:
temp_pair2 = "%d_%s_2.fa" % (i, sample_h)
j = filelib.GenericObject(sample=sample,
pair1=pair1,
pair2=pair2,
temp_pair1=temp_pair1,
temp_pair2=temp_pair2,
bam_filename=bam_filename,
log_filename=log_filename)
jobs.append(j)
for j in jobs:
os.symlink(j.pair1, j.temp_pair1)
if pair2:
os.symlink(j.pair2, j.temp_pair2)
# Generate bowtie2 commands for each of the files.
attr2orient = {
"single": None,
"paired_fr": "fr",
"paired_rf": "rf",
"paired_ff": "ff",
}
orientation = attr2orient[orient.orientation]
#x = sample_node.data.attributes["orientation"]
#orientation = attr2orient[x]
# Takes ~4 Gb per job.
samtools = mlib.findbin("samtools")
sq = parallel.quote
commands = []
for j in jobs:
#sample, pair1, pair2, bam_filename, log_filename = x
nc = max(1, num_cores / len(jobs))
# bowtie2 -p 8 -x <genome> -1 <.fq> -2 <.fq> --fr
# 2> test.log | samtools view -bS -o test.bam -
x1 = alignlib.make_bowtie2_command(ref.fasta_file_full,
j.temp_pair1,
fastq_file2=j.temp_pair2,
orientation=orientation,
num_threads=nc)
x2 = [
sq(samtools),
"view",
"-bS",
"-o",
sq(j.bam_filename),
"-",
]
x2 = " ".join(x2)
x = "%s 2> %s | %s" % (x1, sq(j.log_filename), x2)
#x = "%s >& %s" % (x, sq(log_filename))
commands.append(x)
metadata["commands"] = commands
parallel.pshell(commands, max_procs=num_cores)
# Make sure the analysis completed successfully.
x1 = [x.bam_filename for x in jobs]
x2 = [x.log_filename for x in jobs]
filelib.assert_exists_nz_many(x1 + x2)
return metadata
def run(
self, network, antecedents, out_attributes, user_options, num_cores,
out_path):
import os
from genomicode import parallel
from genomicode import filelib
from genomicode import alignlib
from Betsy import module_utils as mlib
fastq_node, sai_node, orient_node, sample_node, reference_node = \
antecedents
fastq_files = mlib.find_merged_fastq_files(
sample_node.identifier, fastq_node.identifier)
sai_path = sai_node.identifier
assert filelib.dir_exists(sai_path)
orient = mlib.read_orientation(orient_node.identifier)
ref = alignlib.create_reference_genome(reference_node.identifier)
filelib.safe_mkdir(out_path)
metadata = {}
metadata["tool"] = "bwa %s" % alignlib.get_bwa_version()
# Technically, doesn't need the SampleGroupFile, since that's
# already reflected in the sai data. But better, because the
# sai data might not always be generated by BETSY.
# Find the merged fastq files.
# Find the sai files.
sai_filenames = filelib.list_files_in_path(
sai_path, endswith=".sai", case_insensitive=True)
assert sai_filenames, "No .sai files."
bwa = mlib.findbin("bwa")
# bwa samse -f <output.sam> <reference.fa> <input.sai> <input.fq>
# bwa sampe -f <output.sam> <reference.fa> <input_1.sai> <input_2.sai>
# <input_1.fq> <input_2.fq> >
# list of (pair1.fq, pair1.sai, pair2.fq, pair2.sai, output.sam)
# all full paths
jobs = []
for x in fastq_files:
sample, pair1_fq, pair2_fq = x
# The sai file should be in the format:
# <sai_path>/<sample>.sai Single end read
# <sai_path>/<sample>_1.sai Paired end read
# <sai_path>/<sample>_2.sai Paired end read
# Look for pair1_sai and pair2_sai.
pair1_sai = pair2_sai = None
for sai_filename in sai_filenames:
p, s, e = mlib.splitpath(sai_filename)
assert e == ".sai"
if s == sample:
assert not pair1_sai
pair1_sai = sai_filename
elif s == "%s_1" % (sample):
assert not pair1_sai
pair1_sai = sai_filename
elif s == "%s_2" % (sample):
assert not pair2_sai
pair2_sai = sai_filename
assert pair1_sai, "Missing .sai file: %s" % sample
if pair2_fq:
assert pair2_sai, "Missing .sai file 2: %s" % sample
if pair2_sai:
assert pair2_fq, "Missing .fq file 2: %s" % sample
sam_filename = os.path.join(out_path, "%s.sam" % sample)
log_filename = os.path.join(out_path, "%s.log" % sample)
x = sample, pair1_fq, pair1_sai, pair2_fq, pair2_sai, \
sam_filename, log_filename
jobs.append(x)
orientation = orient.orientation
#orientation = sample_node.data.attributes["orientation"]
assert orientation in ["single", "paired_fr", "paired_rf"]
# Make a list of bwa commands.
sq = mlib.sq
commands = []
for x in jobs:
sample, pair1_fq, pair1_sai, pair2_fq, pair2_sai, \
sam_filename, log_filename = x
if orientation == "single":
assert not pair2_fq
assert not pair2_sai
samse = "samse"
if orientation.startswith("paired"):
samse = "sampe"
x = [
sq(bwa),
samse,
"-f", sq(sam_filename),
sq(ref.fasta_file_full),
]
if orientation == "single":
x += [
sq(pair1_sai),
sq(pair1_fq),
]
else:
y = [
sq(pair1_sai),
sq(pair2_sai),
sq(pair1_fq),
sq(pair2_fq),
]
if orientation == "paired_rf":
y = [
sq(pair2_sai),
sq(pair1_sai),
sq(pair2_fq),
sq(pair1_fq),
]
x += y
x += [
">&", sq(log_filename),
]
x = " ".join(x)
commands.append(x)
metadata["commands"] = commands
metadata["num_cores"] = num_cores
parallel.pshell(commands, max_procs=num_cores)
# Make sure the analysis completed successfully.
x = [x[-2] for x in jobs]
filelib.assert_exists_nz_many(x)
return metadata
def run(self, network, antecedents, out_attributes, user_options,
num_cores, out_path):
import os
from genomicode import parallel
from genomicode import filelib
from genomicode import genomelib
from genomicode import config
from Betsy import module_utils as mlib
fasta_node, bam_node, sample_node, orient_node = antecedents
fasta_data = mlib.find_merged_fastq_files(sample_node.identifier,
fasta_node.identifier,
find_fasta=True)
bam_filenames = mlib.find_bam_files(bam_node.identifier)
orient = mlib.read_orientation(orient_node.identifier)
filelib.safe_mkdir(out_path)
# TODO: Try to figure out version.
metadata = {}
metadata["tool"] = "RSeQC (unknown version)"
pyrseqc = mlib.findbin("pyrseqc")
gene_model = mlib.get_user_option(user_options,
"gene_model",
not_empty=True,
allowed_values=["hg19"])
if gene_model == "hg19":
gene_path = config.rseqc_hg19
else:
raise AssertionError, "Unhandled: %s" % gene_model
filelib.dir_exists(gene_path)
gene_model_bed = os.path.join(gene_path, "RefSeq.bed12")
housekeeping_model_bed = os.path.join(gene_path,
"HouseKeepingGenes.bed")
sample2fastadata = {}
for x in fasta_data:
sample, f1, f2 = x
sample2fastadata[sample] = x
is_paired = orient.orientation.startswith("paired")
# Guess the read length. Read the first fasta.
assert sample2fastadata
x = sample2fastadata.keys()[0]
filename = sample2fastadata[x][1]
lengths = {} # length -> count
for i, x in enumerate(genomelib.read_fasta_many(filename)):
if i >= 100:
break
title, sequence = x
l = len(sequence)
lengths[l] = lengths.get(l, 0) + 1
# Use the most common length.
c_length = c_count = None
for (l, c) in lengths.iteritems():
if c_count is None or c > c_count:
c_length, c_count = l, c
assert c_length
read_length = c_length
jobs = [] # sample, bam_filename, fasta_file1, fasta_file2, outdir
for bam_filename in bam_filenames:
# <path>/<sample>.bam
p, sample, e = mlib.splitpath(bam_filename)
assert sample in sample2fastadata
x, f1, f2 = sample2fastadata[sample]
outdir = os.path.join(out_path, sample)
x = sample, bam_filename, f1, f2, outdir
jobs.append(x)
# Some of the modules of RSeQC uses a lot of memory. Have
# seen a Python process take 33 Gb, and an R process take 200
# Gb. However, most of the modules use much less memory. So
# run one pyrseqc at a time, and run each one of those
# processes in parallel. Is probably slower than running
# multiple pyrseqc, but takes less memory.
commands = []
for x in jobs:
sample, bam_filename, fasta_filename1, fasta_filename2, outdir = x
# pyrseqc.py -j 20 --paired_end rqc11.bam rqc14.fa 76 \
# mod07.txt hg19.HouseKeepingGenes.bed rqc21 --dry_run
x = [
mlib.sq(pyrseqc),
"-j",
str(num_cores),
]
if is_paired:
x += ["--paired_end"]
x += [
mlib.sq(bam_filename),
mlib.sq(fasta_filename1),
str(read_length),
mlib.sq(gene_model_bed),
mlib.sq(housekeeping_model_bed),
mlib.sq(outdir),
]
x = " ".join(x)
commands.append(x)
metadata["commands"] = commands
metadata["num_cores"] = num_cores
# pyrseqc takes up to ~40 Gb per process.
# read_distribution.py takes 33 Gb.
# read_quality.py spins off an R process that takes ~200 Gb.
# Make sure we don't use up more memory than is available on
# the machine.
#nc = mlib.calc_max_procs_from_ram(60, upper_max=num_cores)
#metadata["num cores"] = nc
#x = parallel.pshell(commands, max_procs=nc)
# Because of memory, just run one at a time, but each one, use
# multiple cores.
for cmd in commands:
x = parallel.sshell(cmd)
assert x.find("Traceback") < 0, x
filelib.assert_exists_nz(out_path)
return metadata
def merge_vcf_files(vcf_filenames, out_filename, num_cores, tmp_path):
# Put indexed files in tmp_path.
import os
import stat
import shutil
from genomicode import filelib
from genomicode import hashlib
from genomicode import parallel
from Betsy import module_utils as mlib
# TODO: find the version number of these tools.
bgzip = mlib.findbin("bgzip")
tabix = mlib.findbin("tabix")
bcftools = mlib.findbin("bcftools")
sq = parallel.quote
tmp_path = os.path.realpath(tmp_path)
filelib.safe_mkdir(tmp_path)
# Keep track of all commands run.
metadata = {}
metadata["commands"] = []
# Ignore VCF files that don't have any variants.
vcf_filenames = [x for x in vcf_filenames if os.stat(x)[stat.ST_SIZE] > 0]
# If there are no VCF files with any variants, then just create an
# empty outfile and return.
if not vcf_filenames:
open(out_filename, 'w')
return
# 1. Copy VCF files to temporary directory. tmp_filename
# 2. Fix VCF files (e.g. NextGENe, JointSNVMix broken)
# 3. Sort the VCF files (needed for tabix)
# 4. Compress (bgzip)
# 5. Index (tabix)
# 6. Merge
jobs = []
for in_filename in vcf_filenames:
path, root, ext = mlib.splitpath(in_filename)
sample = root
x = "%s%s" % (hashlib.hash_var(root), ext)
tmp_filename = os.path.join(tmp_path, x)
x = filelib.GenericObject(
sample=sample,
in_filename=in_filename,
tmp_filename=tmp_filename,
)
jobs.append(x)
# Make sure temporary files are unique.
seen = {}
for j in jobs:
assert j.tmp_filename not in seen
seen[j.tmp_filename] = 1
# Merge them in order of sample. The germline sample will be
# duplicated, and we will know the order of the germline sample.
schwartz = [(x.sample, x) for x in jobs]
schwartz.sort()
jobs = [x[-1] for x in schwartz]
# Copy all the VCF files to a temporary directory.
for j in jobs:
shutil.copy2(j.in_filename, j.tmp_filename)
#for j in jobs:
# make_file_smaller(j.tmp_filename, 1000)
for j in jobs:
# NextGENe creates broken VCF files. Fix them.
fix_nextgene_vcf(j.tmp_filename)
# JointSNVMix creates broken VCF files. Fix them.
fix_jointsnvmix_vcf(j.tmp_filename)
for j in jobs:
sort_vcf_file(j.tmp_filename)
## # Since we are merging the files, we need to make sure that
## # each file has a unique name. If the names aren't unique,
## # then make them unique by adding the name of the file.
## all_unique = True
## seen = {}
## for x in jobs:
## sample, in_filename, tmp_filename = x
## samples = _get_samples_from_vcf(tmp_filename)
## for s in samples:
## if s in seen:
## all_unique = False
## break
## seen[s] = 1
## if not all_unique:
## break
## if not all_unique:
## for x in jobs:
## sample, in_filename, tmp_filename = x
## _uniquify_samples_in_vcf(tmp_filename, sample)
# Compress the VCF files.
# bgzip file.vcf
commands = []
for j in jobs:
x = "%s %s" % (sq(bgzip), sq(j.tmp_filename))
commands.append(x)
parallel.pshell(commands, max_procs=num_cores, path=tmp_path)
metadata["commands"].extend(commands)
metadata["num_cores"] = num_cores
x = ["%s.gz" % x.tmp_filename for x in jobs]
filelib.assert_exists_nz_many(x)
# Index the VCF files.
# tabix -p vcf file.vcf.gz
commands = []
for j in jobs:
x = "%s -p vcf %s.gz" % (sq(tabix), sq(j.tmp_filename))
commands.append(x)
parallel.pshell(commands, max_procs=num_cores, path=tmp_path)
metadata["commands"].extend(commands)
x = ["%s.gz.tbi" % j.tmp_filename for j in jobs]
filelib.assert_exists_nz_many(x)
# Run bcftools
## For VCF files from somatic calls, the germline sample will
## be duplicated. Add --force-samples to make sure this is
## still merged.
# Since we need to append all the VCF files, it's easy to run
# into error:
# OSError: [Errno 7] Argument list too long
#
# To reduce the chance of this, figure out the path of the
# tmp_filename, and run the analysis in that path so we can
# use relative filenames.
tmp_path = None
for j in jobs:
path, file_ = os.path.split(j.tmp_filename)
if tmp_path is None:
tmp_path = path
assert path == tmp_path
cmd = [
sq(bcftools),
"merge",
"-o %s" % sq(out_filename),
"-O v",
"--force-samples",
]
for j in jobs:
path, file_ = os.path.split(j.tmp_filename)
assert path == tmp_path
cmd.append("%s.gz" % file_)
x = " ".join(cmd)
parallel.sshell(x, path=tmp_path)
metadata["commands"].append(x)
return metadata
def run(self, network, antecedents, out_attributes, user_options,
num_cores, out_path):
import os
import shutil
from genomicode import filelib
from genomicode import parallel
from genomicode import parselib
from Betsy import module_utils as mlib
mpileup_node, nc_node = antecedents
mpileup_filenames = filelib.list_files_in_path(mpileup_node.identifier,
endswith=".pileup")
assert mpileup_filenames, "No .pileup files."
nc_match = mlib.read_normal_cancer_file(nc_node.identifier)
#ref = alignlib.create_reference_genome(ref_node.identifier)
filelib.safe_mkdir(out_path)
# Figure out whether the purpose is to get coverage. Change
# the parameters if it is.
assert "vartype" in out_attributes
vartype = out_attributes["vartype"]
assert vartype in ["snp", "indel"]
sample2pufile = {} # sample -> mpileup filename
for filename in mpileup_filenames:
path, sample, ext = mlib.splitpath(filename)
sample2pufile[sample] = filename
# Make sure files exist for all the samples.
all_samples = []
for (normal_sample, cancer_sample) in nc_match:
if normal_sample not in all_samples:
all_samples.append(normal_sample)
if cancer_sample not in all_samples:
all_samples.append(cancer_sample)
missing = [x for x in all_samples if x not in sample2pufile]
x = parselib.pretty_list(missing, max_items=5)
assert not missing, "Missing BAM files for samples: %s" % x
# list of (sample, normal_pileup, cancer_pileup,
# tmp1_normal, tmp1_cancer, log_filename, out_filename)
opj = os.path.join
jobs = []
for (normal_sample, cancer_sample) in nc_match:
normal_pileup = sample2pufile[normal_sample]
cancer_pileup = sample2pufile[cancer_sample]
p, sample, ext = mlib.splitpath(cancer_pileup)
tmp1_normal = opj(out_path, "%s.normal.tmp1" % sample)
tmp1_cancer = opj(out_path, "%s.cancer.tmp1" % sample)
log_filename = opj(out_path, "%s.log" % sample)
out_filename = opj(out_path, "%s.vcf" % sample)
x = sample, normal_sample, cancer_sample, \
normal_pileup, cancer_pileup, \
tmp1_normal, tmp1_cancer, log_filename, out_filename
jobs.append(x)
# VarScan will generate a "Parsing Exception" if there are 0
# reads in a location. Will be either "0" or blank. Filter
# those lines out.
sq = parallel.quote
commands = []
for x in jobs:
sample, normal_sample, cancer_sample, \
normal_pileup, cancer_pileup, \
tmp1_normal, tmp1_cancer, log_filename, out_filename = x
x1 = "awk -F'\t' '$4 >= 1 {print}' %s > %s" % (normal_pileup,
tmp1_normal)
x2 = "awk -F'\t' '$4 >= 1 {print}' %s > %s" % (cancer_pileup,
tmp1_cancer)
commands.extend([x1, x2])
parallel.pshell(commands, max_procs=num_cores)
x = [x[3] for x in jobs] + [x[4] for x in jobs]
filelib.assert_exists_nz_many(x)
# java -jar VarScan.jar somatic [normal_pileup] [tumor_pileup]
# [output] OPTIONS
varscan = mlib.findbin("varscan_jar")
# Use parameters from:
# Using VarScan 2 for Germline Variant Calling and Somatic
# Mutation Detection
# Make a list of commands.
commands = []
for x in jobs:
sample, normal_sample, cancer_sample, \
normal_pileup, cancer_pileup, \
tmp1_normal, tmp1_cancer, log_filename, out_filename = x
x = [
"java",
"-jar",
sq(varscan),
"somatic",
sq(tmp1_normal),
sq(tmp1_cancer),
sample,
"--min-coverage",
10,
"--min-avg-qual",
15,
"--min-normal-coverage",
10,
"--min-tumor-coverage",
10,
"--min-var-freq",
0.05,
"--somatic-p-value",
0.05,
"--output-vcf",
1,
]
x = " ".join(map(str, x))
x = "%s >& %s" % (x, log_filename)
commands.append(x)
parallel.pshell(commands, max_procs=num_cores)
x = [x[5] for x in jobs]
filelib.assert_exists_nz_many(x)
# Files in out_path can get very big. Clean them up.
# <sample>.normal.tmp1 Very big (10's Gb).
# <sample>.cancer.tmp1 Very big (10's to 100 Gb).
for x in jobs:
sample, normal_sample, cancer_sample, \
normal_pileup, cancer_pileup, \
tmp1_normal, tmp1_cancer, log_filename, out_filename = x
if os.path.exists(tmp1_normal):
os.unlink(tmp1_normal)
if os.path.exists(tmp1_cancer):
os.unlink(tmp1_cancer)
# Copy the final file to the right place.
for x in jobs:
sample, normal_sample, cancer_sample, \
normal_pileup, cancer_pileup, \
tmp1_normal, tmp1_cancer, log_filename, out_filename = x
# Will be written in current directory.
varscan_out = "%s.snp.vcf" % sample
if vartype == "indel":
varscan_out = "%s.indel.vcf" % sample
filelib.assert_exists(varscan_out)
shutil.copy2(varscan_out, out_filename)
# VarScan names the samples "NORMAL" and "TUMOR". Replace
# them with the actual names.
for x in jobs:
sample, normal_sample, cancer_sample, \
normal_pileup, cancer_pileup, \
tmp1_normal, tmp1_cancer, log_filename, out_filename = x
_fix_normal_cancer_names(out_filename, normal_sample,
cancer_sample)
