def extract_rna_info(chrom_info_file, raw_allelic_counts_dir, genotype_dir,time_step, target_regions_dir):
# make dictionary of identifier => index mapping
all_genotype_samples_file = genotype_dir + 'all_genotyped_samples.txt'
samp_idx = get_samples_index(all_genotype_samples_file)
# Initialize chromosome objects
chrom_list = chromosome.get_all_chromosomes(chrom_info_file)
chrom_dict = chromosome.get_chromosome_dict(chrom_info_file)
snp_files = SNPFiles(genotype_dir + 'snp_tab.h5',genotype_dir + 'snp_index.h5',genotype_dir+'haps.h5')
# STEP 1: make combined HDF5 files of AS counts,
# total mapped read counts, and genotype counts
individuals = get_individual_array(target_regions_dir + 'rna_seq_samples_' + str(time_step) + '.txt')
combined_files = CombinedFiles(raw_allelic_counts_dir, chrom_list,time_step)
for ind in individuals:
print(ind)
sample_id = ind + '_' + str(time_step)
count_files = CountFiles(raw_allelic_counts_dir, sample_id)
ind_idx = samp_idx[ind]
combined_files.add_counts(chrom_list, count_files, snp_files, ind_idx)
count_files.close()
return combined_files
def main():
sys.stderr.write("cmd: %s\n" % " ".join(sys.argv))
args = parse_args()
out_f = None
if args.output_file:
if args.output_file.endswith(".gz"):
out_f = gzip.open(args.output_file, "wt")
else:
out_f = open(args.output_file, "wt")
else:
out_f = sys.stdout
# make dictionary of identifier => index mapping
samp_idx = get_samples_index(args)
# read individuals
individuals = read_individuals(args, samp_idx)
chrom_list = chromosome.get_all_chromosomes(args.chrom)
chrom_dict = chromosome.get_chromosome_dict(args.chrom)
combined_files = CombinedFiles(OUTPUT_DIR, chrom_list)
snp_files = SNPFiles(args)
# STEP 1: make combined HDF5 files of AS counts,
# total mapped read counts, and genotype counts
sys.stderr.write("summing genotypes and read counts across individuals\n")
for ind in individuals:
# open count files for this indivudal
sys.stderr.write("individual: %s\n" % ind)
count_files = CountFiles(args.read_count_dir, ind)
ind_idx = samp_idx[ind]
# add counts to combined totals
combined_files.add_counts(chrom_list, count_files, snp_files, ind_idx)
count_files.close()
sys.stderr.write("generating list of target regions\n")
# STEP 2: generate list of target regions centered on test SNPs:
write_target_regions(out_f, args, chrom_list, combined_files, snp_files)
combined_files.close()
snp_files.close()
def main():
sys.stderr.write("cmd: %s\n" % " ".join(sys.argv))
args = parse_args()
out_f = None
if args.output_file:
if args.output_file.endswith(".gz"):
out_f = gzip.open(args.output_file, "wt")
else:
out_f = open(args.output_file, "wt")
else:
out_f = sys.stdout
# make dictionary of identifier => index mapping
samp_idx = get_samples_index(args)
# read individuals
individuals = read_individuals(args, samp_idx)
chrom_list = chromosome.get_all_chromosomes(args.chrom)
chrom_dict = chromosome.get_chromosome_dict(args.chrom)
combined_files = CombinedFiles(OUTPUT_DIR, chrom_list)
snp_files = SNPFiles(args)
# STEP 1: make combined HDF5 files of AS counts,
# total mapped read counts, and genotype counts
sys.stderr.write("summing genotypes and read counts across individuals\n")
for ind in individuals:
# open count files for this indivudal
sys.stderr.write("individual: %s\n" % ind)
count_files = CountFiles(args.read_count_dir, ind)
ind_idx = samp_idx[ind]
# add counts to combined totals
combined_files.add_counts(chrom_list, count_files, snp_files, ind_idx)
count_files.close()
sys.stderr.write("generating list of target regions\n")
# STEP 2: generate list of target regions centered on test SNPs:
write_target_regions(out_f, args, chrom_list, combined_files, snp_files)
combined_files.close()
snp_files.close()
def main():
args = parse_args()
snp_tab_h5 = tables.openFile(args.snp_tab, "r")
snp_index_h5 = tables.openFile(args.snp_index, "r")
if args.haplotype:
hap_h5 = tables.openFile(args.haplotype, "r")
ind_idx = lookup_individual_index(args.samples, args.individual)
else:
hap_h5 = None
ind_idx = None
ref_count_h5 = tables.openFile(args.ref_as_counts, "w")
alt_count_h5 = tables.openFile(args.alt_as_counts, "w")
other_count_h5 = tables.openFile(args.other_as_counts, "w")
read_count_h5 = tables.openFile(args.read_counts, "w")
output_h5 = [ref_count_h5, alt_count_h5, other_count_h5, read_count_h5]
chrom_dict = {}
# initialize every chromosome in output files
chrom_list = chromosome.get_all_chromosomes(args.chrom)
for chrom in chrom_list:
for out_file in output_h5:
create_carray(out_file, chrom, args.data_type)
chrom_dict[chrom.name] = chrom
count = 0
dtype = None
if args.data_type == "uint8":
max_count = MAX_UINT8_COUNT
dtype = np.uint8
elif args.data_type == "uint16":
max_count = MAX_UINT16_COUNT
dtype = np.uint16
else:
raise NotImplementedError("unsupported datatype %s" % args.data_type)
for chrom in chrom_list:
sys.stderr.write("%s\n" % chrom.name)
warned_pos = {}
# fetch SNP info for this chromosome
if chrom.name not in snp_tab_h5.root:
# no SNPs for this chromosome
continue
sys.stderr.write("fetching SNPs\n")
snp_tab = snp_tab_h5.getNode("/%s" % chrom.name)
snp_index_array = snp_index_h5.getNode("/%s" % chrom.name)[:]
if hap_h5:
hap_tab = hap_h5.getNode("/%s" % chrom.name)
else:
hap_tab = None
# initialize count arrays for this chromosome to 0
ref_carray = get_carray(ref_count_h5, chrom)
alt_carray = get_carray(alt_count_h5, chrom)
other_carray = get_carray(other_count_h5, chrom)
read_count_carray = get_carray(read_count_h5, chrom)
ref_array = np.zeros(chrom.length, dtype)
alt_array = np.zeros(chrom.length, dtype)
other_array = np.zeros(chrom.length, dtype)
read_count_array = np.zeros(chrom.length, dtype)
# loop over all BAM files, pulling out reads
# for this chromosome
for bam_filename in args.bam_filenames:
sys.stderr.write("reading from file %s\n" % bam_filename)
samfile = pysam.Samfile(bam_filename, "rb")
for read in get_sam_iter(samfile, chrom):
count += 1
if count == 10000:
sys.stderr.write(".")
count = 0
add_read_count(read, chrom, ref_array, alt_array,
other_array, read_count_array,
snp_index_array, snp_tab, hap_tab,
warned_pos, max_count, ind_idx)
# store results for this chromosome
ref_carray[:] = ref_array
alt_carray[:] = alt_array
other_carray[:] = other_array
read_count_carray[:] = read_count_array
sys.stderr.write("\n")
samfile.close()
# set track statistics and close HDF5 files
sys.stderr.write("setting statistics for each chromosome\n")
for h5f in output_h5:
chromstat.set_stats(h5f, chrom_list)
h5f.close()
snp_tab_h5.close()
snp_index_h5.close()
if hap_h5:
hap_h5.close()
sys.stderr.write("done\n")
def main():
args = parse_args()
sys.stderr.write("command line: %s\n" % " ".join(sys.argv))
sys.stderr.write("python version: %s\n" % sys.version)
sys.stderr.write("pysam version: %s\n" % pysam.__version__)
sys.stderr.write("pytables version: %s\n" % tables.__version__)
util.check_pysam_version()
util.check_pytables_version()
# disable warnings that come from pytables when chromosome
# names are like 1, 2, 3 (instead of chr1, chr2, chr3)
warnings.filterwarnings('ignore', category=tables.NaturalNameWarning)
snp_tab_h5 = tables.open_file(args.snp_tab, "r")
snp_index_h5 = tables.open_file(args.snp_index, "r")
if args.haplotype:
hap_h5 = tables.open_file(args.haplotype, "r")
else:
hap_h5 = None
ref_count_h5 = tables.open_file(args.ref_as_counts, "w")
alt_count_h5 = tables.open_file(args.alt_as_counts, "w")
other_count_h5 = tables.open_file(args.other_as_counts, "w")
read_count_h5 = tables.open_file(args.read_counts, "w")
output_h5 = [ref_count_h5, alt_count_h5, other_count_h5, read_count_h5]
chrom_dict = {}
# initialize every chromosome in output files
chrom_list = chromosome.get_all_chromosomes(args.chrom)
for chrom in chrom_list:
for out_file in output_h5:
create_carray(out_file, chrom, args.data_type)
chrom_dict[chrom.name] = chrom
count = 0
dtype = None
if args.data_type == "uint8":
max_count = MAX_UINT8_COUNT
dtype = np.uint8
elif args.data_type == "uint16":
max_count = MAX_UINT16_COUNT
dtype = np.uint16
else:
raise NotImplementedError("unsupported datatype %s" % args.data_type)
# create a txt file to also holds the counts
if args.txt_counts is not None:
if os.path.splitext(args.txt_counts)[1] == ".gz":
txt_counts = gzip.open(args.txt_counts, 'wt+')
else:
txt_counts = open(args.txt_counts, 'w+')
for chrom in chrom_list:
sys.stderr.write("%s\n" % chrom.name)
if args.test_chrom:
if chrom.name != args.test_chrom:
sys.stderr.write("skipping because not test chrom\n")
continue
warned_pos = {}
# fetch SNP info for this chromosome
if chrom.name not in snp_tab_h5.root:
# no SNPs for this chromosome
sys.stderr.write("skipping %s because chromosome with this name "
"not found in SNP table\n" % chrom.name)
continue
sys.stderr.write("fetching SNPs\n")
snp_tab = snp_tab_h5.get_node("/%s" % chrom.name)
snp_index_array = snp_index_h5.get_node("/%s" % chrom.name)[:]
if hap_h5:
hap_tab = hap_h5.get_node("/%s" % chrom.name)
ind_dict, ind_idx = snptable.SNPTable().get_h5_sample_indices(
hap_h5, chrom, [args.individual])
if len(ind_idx) == 1:
ind_idx = ind_idx[0]
sys.stderr.write("index for individual %s is %d\n" %
(args.individual, ind_idx))
else:
raise ValueError("got sample indices for %d individuals, "
"but expected to get index for one "
"individual (%s)" %
(len(ind_idx), args.individual))
hap_tab = None
ind_idx = None
else:
hap_tab = None
ind_idx = None
# initialize count arrays for this chromosome to 0
ref_carray = get_carray(ref_count_h5, chrom)
alt_carray = get_carray(alt_count_h5, chrom)
other_carray = get_carray(other_count_h5, chrom)
read_count_carray = get_carray(read_count_h5, chrom)
ref_array = np.zeros(chrom.length, dtype)
alt_array = np.zeros(chrom.length, dtype)
other_array = np.zeros(chrom.length, dtype)
read_count_array = np.zeros(chrom.length, dtype)
# loop over all BAM files, pulling out reads
# for this chromosome
for bam_filename in args.bam_filenames:
sys.stderr.write("reading from file %s\n" % bam_filename)
samfile = pysam.Samfile(bam_filename, "rb")
for read in get_sam_iter(samfile, chrom):
count += 1
if count == 10000:
sys.stderr.write(".")
count = 0
add_read_count(read, chrom, ref_array, alt_array, other_array,
read_count_array, snp_index_array, snp_tab,
hap_tab, warned_pos, max_count, ind_idx)
# store results for this chromosome
ref_carray[:] = ref_array
alt_carray[:] = alt_array
other_carray[:] = other_array
read_count_carray[:] = read_count_array
sys.stderr.write("\n")
# write data to numpy arrays, so that they can be written to a txt
# file later
# columns are:
# chrom, pos, ref, alt, genotype, ref_count, alt_count, other_count
if args.txt_counts is not None:
write_txt_file(txt_counts, chrom, snp_tab, hap_tab, ind_idx,
ref_array, alt_array, other_array)
samfile.close()
if args.txt_counts:
# close the open txt file handler
txt_counts.close()
# check if any of the reads contained an unimplemented CIGAR
if unimplemented_CIGAR[0] > 0:
sys.stderr.write("WARNING: Encountered " +
str(unimplemented_CIGAR[0]) +
" instances of CIGAR codes: " +
str(unimplemented_CIGAR[1]) + ". Reads with these "
"CIGAR codes were skipped because they "
"are currently unimplemented.\n")
# set track statistics and close HDF5 files
sys.stderr.write("setting statistics for each chromosome\n")
for h5f in output_h5:
chromstat.set_stats(h5f, chrom_list)
h5f.close()
snp_tab_h5.close()
snp_index_h5.close()
if hap_h5:
hap_h5.close()
sys.stderr.write("done\n")
def main():
args = parse_args()
write_header(sys.stdout)
# find index of individual in list of samples
ind_idx = lookup_individual_index(args, args.individual)
data_files = DataFiles(args)
chrom_list = chromosome.get_all_chromosomes(args.chrom)
chrom_dict = chromosome.get_chromosome_dict(args.chrom)
genomewide_read_counts = get_genomewide_count(data_files.read_count_h5,
chrom_list)
if args.input_file.endswith(".gz"):
f = gzip.open(args.input_file)
else:
f = open(args.input_file)
line_count = 0
if args.target_region_size:
sys.stderr.write("setting target region size to %d\n" %
args.target_region_size)
for line in f:
line_count += 1
if line_count % 1000 == 0:
sys.stderr.write(".")
if line.startswith("#"):
continue
words = line.rstrip().split()
if words[1] == "NA":
# no SNP defined on this line:
write_NA_line(sys.stdout)
continue
chrom_name = words[0]
chrom = chrom_dict[chrom_name]
region_list = get_target_regions(args, chrom, words)
snp_pos = int(words[1])
snp_ref_base = words[3]
snp_alt_base = words[4]
# TODO: check that SNP ref/alt match?
snp_region = coord.Coord(chrom, snp_pos, snp_pos)
# pull out all of the SNPs in the target region(s)
region_snps = get_region_snps(data_files, region_list, ind_idx)
# pull out test SNP
test_snp_list = get_region_snps(data_files, [snp_region], ind_idx)
if len(test_snp_list) != 1:
test_snp = None
sys.stderr.write("WARNING: could not find test SNP at "
"position %s:%d\n" % (chrom.name, snp_pos))
het_snps = []
else:
test_snp = test_snp_list[0]
# pull out haplotype counts from linked heterozygous SNPs
het_snps = get_het_snps(region_snps)
set_snp_counts(data_files, region_list, het_snps, test_snp, args)
region_read_counts = get_region_read_counts(data_files, region_list)
write_output(sys.stdout, region_list, het_snps, test_snp, snp_pos,
region_read_counts, genomewide_read_counts)
sys.stderr.write("\n")
f.close()
data_files.close()
def main():
args = parse_args()
sys.stderr.write("command line: %s\n" % " ".join(sys.argv))
sys.stderr.write("python version: %s\n" % sys.version)
sys.stderr.write("pysam version: %s\n" % pysam.__version__)
sys.stderr.write("pytables version: %s\n" % tables.__version__)
util.check_pysam_version()
util.check_pytables_version()
# disable warnings that come from pytables when chromosome
# names are like 1, 2, 3 (instead of chr1, chr2, chr3)
warnings.filterwarnings('ignore', category=tables.NaturalNameWarning)
snp_tab_h5 = tables.open_file(args.snp_tab, "r")
snp_index_h5 = tables.open_file(args.snp_index, "r")
if args.haplotype:
hap_h5 = tables.open_file(args.haplotype, "r")
else:
hap_h5 = None
ref_count_h5 = tables.open_file(args.ref_as_counts, "w")
alt_count_h5 = tables.open_file(args.alt_as_counts, "w")
other_count_h5 = tables.open_file(args.other_as_counts, "w")
read_count_h5 = tables.open_file(args.read_counts, "w")
output_h5 = [ref_count_h5, alt_count_h5, other_count_h5, read_count_h5]
chrom_dict = {}
# initialize every chromosome in output files
chrom_list = chromosome.get_all_chromosomes(args.chrom)
for chrom in chrom_list:
for out_file in output_h5:
create_carray(out_file, chrom, args.data_type)
chrom_dict[chrom.name] = chrom
count = 0
dtype = None
if args.data_type == "uint8":
max_count = MAX_UINT8_COUNT
dtype = np.uint8
elif args.data_type == "uint16":
max_count = MAX_UINT16_COUNT
dtype = np.uint16
else:
raise NotImplementedError("unsupported datatype %s" % args.data_type)
# create a txt file to also holds the counts
if args.txt_counts is not None:
if os.path.splitext(args.txt_counts)[1] == ".gz":
txt_counts = gzip.open(args.txt_counts, 'wt+')
else:
txt_counts = open(args.txt_counts, 'w+')
for chrom in chrom_list:
sys.stderr.write("%s\n" % chrom.name)
if args.test_chrom:
if chrom.name != args.test_chrom:
sys.stderr.write("skipping because not test chrom\n")
continue
warned_pos = {}
# fetch SNP info for this chromosome
if chrom.name not in snp_tab_h5.root:
# no SNPs for this chromosome
sys.stderr.write("skipping %s because chromosome with this name "
"not found in SNP table\n" % chrom.name)
continue
sys.stderr.write("fetching SNPs\n")
snp_tab = snp_tab_h5.get_node("/%s" % chrom.name)
snp_index_array = snp_index_h5.get_node("/%s" % chrom.name)[:]
if hap_h5:
hap_tab = hap_h5.get_node("/%s" % chrom.name)
ind_dict, ind_idx = snptable.SNPTable().get_h5_sample_indices(
hap_h5, chrom, [args.individual])
if len(ind_idx) == 1:
ind_idx = ind_idx[0]
sys.stderr.write("index for individual %s is %d\n" %
(args.individual, ind_idx))
else:
raise ValueError("got sample indices for %d individuals, "
"but expected to get index for one "
"individual (%s)" % (len(ind_idx),
args.individual))
hap_tab = None
ind_idx = None
else:
hap_tab = None
ind_idx = None
# initialize count arrays for this chromosome to 0
ref_carray = get_carray(ref_count_h5, chrom)
alt_carray = get_carray(alt_count_h5, chrom)
other_carray = get_carray(other_count_h5, chrom)
read_count_carray = get_carray(read_count_h5, chrom)
ref_array = np.zeros(chrom.length, dtype)
alt_array = np.zeros(chrom.length, dtype)
other_array = np.zeros(chrom.length, dtype)
read_count_array = np.zeros(chrom.length, dtype)
# loop over all BAM files, pulling out reads
# for this chromosome
for bam_filename in args.bam_filenames:
sys.stderr.write("reading from file %s\n" % bam_filename)
samfile = pysam.Samfile(bam_filename, "rb")
for read in get_sam_iter(samfile, chrom):
count += 1
if count == 10000:
sys.stderr.write(".")
count = 0
add_read_count(read, chrom, ref_array, alt_array,
other_array, read_count_array,
snp_index_array, snp_tab, hap_tab,
warned_pos, max_count, ind_idx)
# store results for this chromosome
ref_carray[:] = ref_array
alt_carray[:] = alt_array
other_carray[:] = other_array
read_count_carray[:] = read_count_array
sys.stderr.write("\n")
# write data to numpy arrays, so that they can be written to a txt
# file later
# columns are:
# chrom, pos, ref, alt, genotype, ref_count, alt_count, other_count
if args.txt_counts is not None:
write_txt_file(txt_counts, chrom, snp_tab, hap_tab, ind_idx,
ref_array, alt_array, other_array)
samfile.close()
if args.txt_counts:
# close the open txt file handler
txt_counts.close()
# check if any of the reads contained an unimplemented CIGAR
if unimplemented_CIGAR[0] > 0:
sys.stderr.write("WARNING: Encountered " + str(unimplemented_CIGAR[0])
+ " instances of CIGAR codes: "
+ str(unimplemented_CIGAR[1]) + ". Reads with these "
"CIGAR codes were skipped because they "
"are currently unimplemented.\n")
# set track statistics and close HDF5 files
sys.stderr.write("setting statistics for each chromosome\n")
for h5f in output_h5:
chromstat.set_stats(h5f, chrom_list)
h5f.close()
snp_tab_h5.close()
snp_index_h5.close()
if hap_h5:
hap_h5.close()
sys.stderr.write("done\n")
def main():
args = parse_args()
sys.stderr.write("command line: %s\n" % " ".join(sys.argv))
sys.stderr.write("python version: %s\n" % sys.version)
sys.stderr.write("pysam version: %s\n" % pysam.__version__)
sys.stderr.write("pytables version: %s\n" % tables.__version__)
util.check_pysam_version()
util.check_pytables_version()
snp_tab_h5 = tables.open_file(args.snp_tab, "r")
snp_index_h5 = tables.open_file(args.snp_index, "r")
if args.haplotype:
hap_h5 = tables.open_file(args.haplotype, "r")
ind_idx = lookup_individual_index(args.samples, args.individual)
else:
hap_h5 = None
ind_idx = None
ref_count_h5 = tables.open_file(args.ref_as_counts, "w")
alt_count_h5 = tables.open_file(args.alt_as_counts, "w")
other_count_h5 = tables.open_file(args.other_as_counts, "w")
read_count_h5 = tables.open_file(args.read_counts, "w")
output_h5 = [ref_count_h5, alt_count_h5, other_count_h5, read_count_h5]
chrom_dict = {}
# initialize every chromosome in output files
chrom_list = chromosome.get_all_chromosomes(args.chrom)
for chrom in chrom_list:
for out_file in output_h5:
create_carray(out_file, chrom, args.data_type)
chrom_dict[chrom.name] = chrom
count = 0
dtype = None
if args.data_type == "uint8":
max_count = MAX_UINT8_COUNT
dtype = np.uint8
elif args.data_type == "uint16":
max_count = MAX_UINT16_COUNT
dtype = np.uint16
else:
raise NotImplementedError("unsupported datatype %s" % args.data_type)
# create a list to hold the counts that will be later written
# to a txt file
if args.text_counts is not None:
txt_counts = list()
for chrom in chrom_list:
sys.stderr.write("%s\n" % chrom.name)
warned_pos = {}
# fetch SNP info for this chromosome
if chrom.name not in snp_tab_h5.root:
# no SNPs for this chromosome
continue
sys.stderr.write("fetching SNPs\n")
snp_tab = snp_tab_h5.get_node("/%s" % chrom.name)
snp_index_array = snp_index_h5.get_node("/%s" % chrom.name)[:]
if hap_h5:
hap_tab = hap_h5.get_node("/%s" % chrom.name)
else:
hap_tab = None
# initialize count arrays for this chromosome to 0
ref_carray = get_carray(ref_count_h5, chrom)
alt_carray = get_carray(alt_count_h5, chrom)
other_carray = get_carray(other_count_h5, chrom)
read_count_carray = get_carray(read_count_h5, chrom)
ref_array = np.zeros(chrom.length, dtype)
alt_array = np.zeros(chrom.length, dtype)
other_array = np.zeros(chrom.length, dtype)
read_count_array = np.zeros(chrom.length, dtype)
# loop over all BAM files, pulling out reads
# for this chromosome
for bam_filename in args.bam_filenames:
sys.stderr.write("reading from file %s\n" % bam_filename)
samfile = pysam.Samfile(bam_filename, "rb")
for read in get_sam_iter(samfile, chrom):
count += 1
if count == 10000:
sys.stderr.write(".")
count = 0
add_read_count(read, chrom, ref_array, alt_array,
other_array, read_count_array,
snp_index_array, snp_tab, hap_tab,
warned_pos, max_count, ind_idx)
# store results for this chromosome
ref_carray[:] = ref_array
alt_carray[:] = alt_array
other_carray[:] = other_array
read_count_carray[:] = read_count_array
sys.stderr.write("\n")
# write data to numpy arrays, so that they can be written to a txt
# file later
# columns are:
# chrom, pos, ref, alt, genotype, ref_count, alt_count, other_count
if args.text_counts is not None:
chrom = np.tile(chrom.name, len(snp_tab))
pos = np.array([snp['pos'] for snp in snp_tab])
ref = np.array([snp['allele1'] for snp in snp_tab])
alt = np.array([snp['allele2'] for snp in snp_tab])
if hap_tab is not None:
genotype = np.array([str(hap[0])+"|"+str(hap[1])
for hap in hap_tab])
else:
genotype = np.empty((len(snp_tab), 0))
txt_counts.append(
np.column_stack((chrom, pos, ref, alt, genotype,
ref_array[pos-1],
alt_array[pos-1],
other_array[pos-1]))
)
samfile.close()
# write the txt_counts np arrays to a txt file
if args.text_counts is not None:
# we use vstack to combine np arrays row-wise into a multi-dimensional
# array
np.savetxt(args.text_counts, np.vstack(tuple(txt_counts)),
fmt="%1s", delimiter=" ")
# set track statistics and close HDF5 files
sys.stderr.write("setting statistics for each chromosome\n")
for h5f in output_h5:
chromstat.set_stats(h5f, chrom_list)
h5f.close()
snp_tab_h5.close()
snp_index_h5.close()
if hap_h5:
hap_h5.close()
sys.stderr.write("done\n")
def main():
args = parse_args()
write_header(sys.stdout)
# find index of individual in list of samples
ind_idx = lookup_individual_index(args, args.individual)
data_files = DataFiles(args)
chrom_list = chromosome.get_all_chromosomes(args.chrom)
chrom_dict = chromosome.get_chromosome_dict(args.chrom)
genomewide_read_counts = get_genomewide_count(data_files.read_count_h5,
chrom_list)
unknown_chrom = set([])
if util.is_gzipped(args.input_file):
f = gzip.open(args.input_file, "rt")
else:
f = open(args.input_file, "r")
line_count = 0
if args.target_region_size:
sys.stderr.write("setting target region size to %d\n" %
args.target_region_size)
for line in f:
line_count += 1
if line_count % 1000 == 0:
sys.stderr.write(".")
if line.startswith("#"):
continue
words = line.rstrip().split()
if words[1] == "NA":
# no SNP defined on this line:
write_NA_line(sys.stdout)
continue
chrom_name = words[0]
if chrom_name in chrom_dict:
chrom = chrom_dict[chrom_name]
else:
if not chrom_name.startswith("chr"):
# try adding 'chr' to front of name
new_chrom_name = "chr" + chrom_name
if new_chrom_name in chrom_dict:
chrom_name = new_chrom_name
chrom = chrom_dict[chrom_name]
else:
# can't figure out this chromosome name
if not chrom_name in unknown_chrom:
unknown_chrom.add(chrom_name)
sys.stderr.write("WARNING: unknown chromosome '%s'")
continue
region_list = get_target_regions(args, chrom, words)
snp_pos = int(words[1])
snp_ref_base = words[3]
snp_alt_base = words[4]
# TODO: check that SNP ref/alt match?
snp_region = coord.Coord(chrom, snp_pos, snp_pos)
# pull out all of the SNPs in the target region(s)
region_snps = get_region_snps(data_files, region_list, ind_idx)
# pull out test SNP
test_snp_list = get_region_snps(data_files, [snp_region], ind_idx)
if len(test_snp_list) != 1:
test_snp = None
sys.stderr.write("WARNING: could not find test SNP at "
"position %s:%d\n" % (chrom.name, snp_pos))
het_snps = []
else:
test_snp = test_snp_list[0]
# pull out haplotype counts from linked heterozygous SNPs
het_snps = get_het_snps(region_snps)
set_snp_counts(data_files, region_list, het_snps, test_snp, args)
region_read_counts = get_region_read_counts(data_files, region_list)
write_output(sys.stdout, region_list, het_snps, test_snp, snp_pos,
region_read_counts, genomewide_read_counts)
sys.stderr.write("\n")
f.close()
data_files.close()
def main():
args = parse_args()
sys.stderr.write("command line: %s\n" % " ".join(sys.argv))
sys.stderr.write("python version: %s\n" % sys.version)
sys.stderr.write("pysam version: %s\n" % pysam.__version__)
sys.stderr.write("pytables version: %s\n" % tables.__version__)
util.check_pysam_version()
util.check_pytables_version()
snp_tab_h5 = tables.open_file(args.snp_tab, "r")
snp_index_h5 = tables.open_file(args.snp_index, "r")
if args.haplotype:
hap_h5 = tables.open_file(args.haplotype, "r")
else:
hap_h5 = None
ref_count_h5 = tables.open_file(args.ref_as_counts, "w")
alt_count_h5 = tables.open_file(args.alt_as_counts, "w")
other_count_h5 = tables.open_file(args.other_as_counts, "w")
read_count_h5 = tables.open_file(args.read_counts, "w")
output_h5 = [ref_count_h5, alt_count_h5, other_count_h5, read_count_h5]
chrom_dict = {}
# initialize every chromosome in output files
chrom_list = chromosome.get_all_chromosomes(args.chrom)
for chrom in chrom_list:
for out_file in output_h5:
create_carray(out_file, chrom, args.data_type)
chrom_dict[chrom.name] = chrom
count = 0
dtype = None
if args.data_type == "uint8":
max_count = MAX_UINT8_COUNT
dtype = np.uint8
elif args.data_type == "uint16":
max_count = MAX_UINT16_COUNT
dtype = np.uint16
else:
raise NotImplementedError("unsupported datatype %s" % args.data_type)
# create a txt file to also holds the counts
if args.txt_counts is not None:
if os.path.splitext(args.txt_counts)[1] == ".gz":
txt_counts = gzip.open(args.txt_counts, 'a+')
else:
txt_counts = open(args.txt_counts, 'a+')
for chrom in chrom_list:
sys.stderr.write("%s\n" % chrom.name)
warned_pos = {}
# fetch SNP info for this chromosome
if chrom.name not in snp_tab_h5.root:
# no SNPs for this chromosome
continue
sys.stderr.write("fetching SNPs\n")
snp_tab = snp_tab_h5.get_node("/%s" % chrom.name)
snp_index_array = snp_index_h5.get_node("/%s" % chrom.name)[:]
if hap_h5:
hap_tab = hap_h5.get_node("/%s" % chrom.name)
ind_idx = snptable.SNPTable().get_h5_sample_indices(
hap_h5, chrom, [args.individual])[1]
if len(ind_idx) != 0:
ind_idx = ind_idx[0]
else:
hap_tab = None
ind_idx = None
else:
hap_tab = None
ind_idx = None
# initialize count arrays for this chromosome to 0
ref_carray = get_carray(ref_count_h5, chrom)
alt_carray = get_carray(alt_count_h5, chrom)
other_carray = get_carray(other_count_h5, chrom)
read_count_carray = get_carray(read_count_h5, chrom)
ref_array = np.zeros(chrom.length, dtype)
alt_array = np.zeros(chrom.length, dtype)
other_array = np.zeros(chrom.length, dtype)
read_count_array = np.zeros(chrom.length, dtype)
# loop over all BAM files, pulling out reads
# for this chromosome
for bam_filename in args.bam_filenames:
sys.stderr.write("reading from file %s\n" % bam_filename)
samfile = pysam.Samfile(bam_filename, "rb")
for read in get_sam_iter(samfile, chrom):
count += 1
if count == 10000:
sys.stderr.write(".")
count = 0
add_read_count(read, chrom, ref_array, alt_array, other_array,
read_count_array, snp_index_array, snp_tab,
hap_tab, warned_pos, max_count, ind_idx)
# store results for this chromosome
ref_carray[:] = ref_array
alt_carray[:] = alt_array
other_carray[:] = other_array
read_count_carray[:] = read_count_array
sys.stderr.write("\n")
# write data to numpy arrays, so that they can be written to a txt
# file later
# columns are:
# chrom, pos, ref, alt, genotype, ref_count, alt_count, other_count
if args.txt_counts is not None:
chrom = np.tile(chrom.name, len(snp_tab))
pos = np.array([snp['pos'] for snp in snp_tab])
ref = np.array([snp['allele1'] for snp in snp_tab])
alt = np.array([snp['allele2'] for snp in snp_tab])
if hap_tab is not None:
genotype = np.array(
[str(hap[0]) + "|" + str(hap[1]) for hap in hap_tab])
else:
genotype = np.empty((len(snp_tab), 0))
# write an np array to a txt file
np.savetxt(txt_counts,
np.column_stack(
(chrom, pos, ref, alt, genotype,
ref_array[pos - 1], alt_array[pos - 1],
other_array[pos - 1])),
fmt="%1s",
delimiter=" ")
samfile.close()
if args.txt_counts:
# close the open txt file handler
txt_counts.close()
# check if any of the reads contained an unimplemented CIGAR
sys.stderr.write(
"WARNING: Encountered " + str(unimplemented_CIGAR[0]) +
" instances of any of the following CIGAR codes: " +
str(unimplemented_CIGAR[1]) +
". The regions of reads with these CIGAR codes were skipped because these CIGAR codes are currently unimplemented.\n"
)
# set track statistics and close HDF5 files
sys.stderr.write("setting statistics for each chromosome\n")
for h5f in output_h5:
chromstat.set_stats(h5f, chrom_list)
h5f.close()
snp_tab_h5.close()
snp_index_h5.close()
if hap_h5:
hap_h5.close()
sys.stderr.write("done\n")
def main():
args = parse_args()
snp_tab_h5 = tables.openFile(args.snp_tab, "r")
snp_index_h5 = tables.openFile(args.snp_index, "r")
if args.haplotype:
hap_h5 = tables.openFile(args.haplotype, "r")
ind_idx = lookup_individual_index(args.samples, args.individual)
else:
hap_h5 = None
ind_idx = None
ref_count_h5 = tables.openFile(args.ref_as_counts, "w")
alt_count_h5 = tables.openFile(args.alt_as_counts, "w")
other_count_h5 = tables.openFile(args.other_as_counts, "w")
read_count_h5 = tables.openFile(args.read_counts, "w")
output_h5 = [ref_count_h5, alt_count_h5, other_count_h5, read_count_h5]
chrom_dict = {}
# initialize every chromosome in output files
chrom_list = chromosome.get_all_chromosomes(args.chrom)
for chrom in chrom_list:
for out_file in output_h5:
create_carray(out_file, chrom, args.data_type)
chrom_dict[chrom.name] = chrom
count = 0
dtype = None
if args.data_type == "uint8":
max_count = MAX_UINT8_COUNT
dtype = np.uint8
elif args.data_type == "uint16":
max_count = MAX_UINT16_COUNT
dtype = np.uint16
else:
raise NotImplementedError("unsupported datatype %s" % args.data_type)
for chrom in chrom_list:
sys.stderr.write("%s\n" % chrom.name)
warned_pos = {}
# fetch SNP info for this chromosome
if chrom.name not in snp_tab_h5.root:
# no SNPs for this chromosome
continue
sys.stderr.write("fetching SNPs\n")
snp_tab = snp_tab_h5.getNode("/%s" % chrom.name)
snp_index_array = snp_index_h5.getNode("/%s" % chrom.name)[:]
if hap_h5:
hap_tab = hap_h5.getNode("/%s" % chrom.name)
else:
hap_tab = None
# initialize count arrays for this chromosome to 0
ref_carray = get_carray(ref_count_h5, chrom)
alt_carray = get_carray(alt_count_h5, chrom)
other_carray = get_carray(other_count_h5, chrom)
read_count_carray = get_carray(read_count_h5, chrom)
ref_array = np.zeros(chrom.length, dtype)
alt_array = np.zeros(chrom.length, dtype)
other_array = np.zeros(chrom.length, dtype)
read_count_array = np.zeros(chrom.length, dtype)
# loop over all BAM files, pulling out reads
# for this chromosome
for bam_filename in args.bam_filenames:
sys.stderr.write("reading from file %s\n" % bam_filename)
samfile = pysam.Samfile(bam_filename, "rb")
for read in get_sam_iter(samfile, chrom):
count += 1
if count == 10000:
sys.stderr.write(".")
count = 0
add_read_count(read, chrom, ref_array, alt_array, other_array,
read_count_array, snp_index_array, snp_tab,
hap_tab, warned_pos, max_count, ind_idx)
# store results for this chromosome
ref_carray[:] = ref_array
alt_carray[:] = alt_array
other_carray[:] = other_array
read_count_carray[:] = read_count_array
sys.stderr.write("\n")
samfile.close()
# set track statistics and close HDF5 files
sys.stderr.write("setting statistics for each chromosome\n")
for h5f in output_h5:
chromstat.set_stats(h5f, chrom_list)
h5f.close()
snp_tab_h5.close()
snp_index_h5.close()
if hap_h5:
hap_h5.close()
sys.stderr.write("done\n")
