def extract_annotation_by_refence_id(list_of_target_gff, id_file,
extracted_gff, filtered_out_gff):
ids = IdList()
ids.read(id_file)
extracted_gff_fd = open(extracted_gff, "w")
filtered_out_gff_fd = open(filtered_out_gff, "w")
for filename in list_of_target_gff:
with open(filename, "r") as in_fd:
for line in in_fd:
tmp = line
if tmp == "# --- START OF GFF DUMP ---\n":
# read until string with target_name will appear
while tmp[0] == "#":
tmp = next(in_fd, "")
target_name = tmp.split("\t")[8].split(
";")[1].split()[1]
if target_name not in ids:
writing_fd = filtered_out_gff_fd
else:
writing_fd = extracted_gff_fd
# print target_name
writing_fd.write(tmp)
while True:
tmp = next(in_fd, "")
if tmp == "# --- END OF GFF DUMP ---\n":
break
writing_fd.write(tmp)
if tmp == "":
break
extracted_gff_fd.close()
filtered_out_gff_fd.close()
def extract_top_hits_from_target_gff(list_of_target_gff,
top_hits_gff,
secondary_hits_gff,
id_white_list_file=None,
max_hits_per_query=None):
if id_white_list_file:
white_ids = IdList()
white_ids.read(id_white_list_file)
top_hits_gff_fd = open(top_hits_gff, "w")
secondary_hits_gff_fd = open(secondary_hits_gff, "w")
targets_list = []
hit_counter = 0
gene_counter = 0
for filename in list_of_target_gff:
index = 0
with open(filename, "r") as in_fd:
#print u
#tmp = None
for line in in_fd:
tmp = line
if tmp == "# --- START OF GFF DUMP ---\n":
# read until string with target_name will appear
while tmp[0] == "#":
tmp = next(in_fd, "")
target_name = tmp.split("\t")[8].split(
";")[1].split()[1]
if id_white_list_file:
if target_name not in white_ids:
continue
if target_name not in targets_list:
writing_fd = top_hits_gff_fd
targets_list.append(target_name)
gene_counter += 1
hit_counter = 0
else:
writing_fd = secondary_hits_gff_fd
# print target_name
hit_counter += 1
tmp = tmp.replace(
"gene_id 0",
"gene_id g%i_h%i" % (gene_counter, hit_counter))
if hit_counter <= max_hits_per_query:
writing_fd.write(tmp)
while True:
tmp = next(in_fd, "")
# print("cccc")
if tmp == "# --- END OF GFF DUMP ---\n":
break
if max_hits_per_query:
if hit_counter > max_hits_per_query:
#print "aaaaa"
continue
writing_fd.write(tmp)
if tmp == "":
break
top_hits_gff_fd.close()
secondary_hits_gff_fd.close()
def extract_sequences_from_selected_clusters(
self,
clusters_id_file,
cluster_file,
seq_file,
output_dir="./",
seq_format="fasta",
out_prefix=None,
create_dir_for_each_cluster=False,
skip_cluster_if_no_sequence_for_element=True):
from Routines import SequenceRoutines, FileRoutines
cluster_id_list = IdList()
cluster_dict = SynDict()
#print(pep_file)
FileRoutines.safe_mkdir(output_dir)
out_dir = FileRoutines.check_path(output_dir)
create_directory_for_each_cluster = True if out_prefix else create_dir_for_each_cluster
if clusters_id_file:
cluster_id_list.read(clusters_id_file)
cluster_dict.read(cluster_file,
split_values=True,
values_separator=",")
protein_dict = SeqIO.index_db(
"tmp.idx",
FileRoutines.make_list_of_path_to_files(seq_file),
format=seq_format)
number_of_skipped_clusters = 0
for fam_id in cluster_id_list if clusters_id_file else cluster_dict:
if skip_cluster_if_no_sequence_for_element:
absent_elements = self.check_absence_of_cluster_elements(
cluster_dict[fam_id], protein_dict)
if absent_elements:
print "Skipping cluster %s due to absent element(%s)" % (
fam_id, ",".join(absent_elements))
number_of_skipped_clusters += 1
continue
if fam_id in cluster_dict:
if create_directory_for_each_cluster:
fam_dir = "%s%s/" % (out_dir, fam_id)
FileRoutines.safe_mkdir(fam_dir)
out_file = "%s%s.fasta" % (fam_dir, out_prefix
if out_prefix else fam_id)
else:
out_file = "%s/%s.fasta" % (out_dir, out_prefix
if out_prefix else fam_id)
SeqIO.write(SequenceRoutines.record_by_id_generator(
protein_dict, cluster_dict[fam_id], verbose=True),
out_file,
format=seq_format)
os.remove("tmp.idx")
print "%i of %i clusters were skipped due to absent elements" % (
number_of_skipped_clusters, len(cluster_dict))
return number_of_skipped_clusters
def extract_ids_from_file(input_file, output_file=None, header=False, column_separator="\t",
comments_prefix="#", column_number=None):
id_list = IdList()
id_list.read(input_file, column_separator=column_separator, comments_prefix=comments_prefix,
column_number=column_number, header=header)
if output_file:
id_list.write(output_file, header=header)
return id_list
def split_hmm(self, hmmfile, output_dir, num_of_recs_per_file, num_of_files=None, output_prefix=None, threads=4):
try:
os.mkdir(output_dir)
except OSError:
pass
id_fd = CGAS.cgas(hmmfile, grep_pattern="NAME", whole_word_match=True, awk_code="{print $2}",
capture_output=True)
split_index = 1
ids_written = 0
ids_list = IdList()
#ids_list = read_ids(id_fd, close_after_if_file_object=False)
ids_list.read(id_fd, close_after_if_file_object=True)
number_of_ids = len(ids_list)
out_prefix = self.split_filename(hmmfile)[1] if output_prefix is None else output_prefix
num_of_ids = int(number_of_ids/num_of_files) + 1 if num_of_files else num_of_recs_per_file
common_options = " -f"
common_options += " %s" % hmmfile
options_list = []
while (ids_written + num_of_ids) <= number_of_ids:
tmp_id_list = IdList(ids_list[ids_written:ids_written+num_of_ids])
tmp_id_list.write("%s/%s_%i.ids" % (output_dir, out_prefix, split_index))
options = common_options
options += " %s/%s_%i.ids" % (output_dir, out_prefix, split_index)
options += " > %s" % ("%s/%s_%i.hmm" % (output_dir, out_prefix, split_index))
options_list.append(options)
split_index += 1
ids_written += num_of_ids
if ids_written != number_of_ids:
tmp_id_list = IdList(ids_list[ids_written:])
tmp_id_list.write("%s/%s_%i.ids" % (output_dir, out_prefix, split_index))
options = common_options
options += " %s/%s_%i.ids" % (output_dir, out_prefix, split_index)
options += " > %s" % ("%s/%s_%i.hmm" % (output_dir, out_prefix, split_index))
options_list.append(options)
split_index += 1
#print options_list
self.parallel_execute(options_list, cmd="hmmfetch", threads=threads)
def extract_clusters_by_element_ids_from_file(self,
cluster_file,
element_file,
output_file,
mode="w"):
""""
mode: "w" - if elements from element_id_list are present in cluster extracts only that elements
"a" - if elements from element_id_list are present in cluster extracts all elements
"""
cluster_dict = SynDict()
cluster_dict.read(cluster_file, split_values=True, comments_prefix="#")
element_id_list = IdList()
element_id_list.read(element_file, comments_prefix="#")
extracted_clusters = self.extract_clusters_by_element_ids(
cluster_dict, element_id_list, mode=mode)
extracted_clusters.write(output_file, splited_values=True)
def extract_evidence_by_ids(evidence_file, id_file, output_evidence_file, mode="transcript"):
# possible modes: transcript, gene
ids = IdList()
ids.read(id_file, comments_prefix="#")
column_id = 0 if mode == "gene" else 1
with open(evidence_file, "r") as ev_fd:
with open(output_evidence_file, "w") as out_fd:
for line in ev_fd:
if line[0] == "#":
out_fd.write(line)
continue
entry_id = line.split("\t")[column_id]
if entry_id in ids:
out_fd.write(line)
def extract_proteins_from_selected_families(
families_id_file,
fam_file,
pep_file,
output_dir="./",
pep_format="fasta",
out_prefix=None,
create_dir_for_each_family=False):
from Routines import SequenceRoutines, FileRoutines
fam_id_list = IdList()
fam_dict = SynDict()
#print(pep_file)
FileRoutines.safe_mkdir(output_dir)
out_dir = FileRoutines.check_path(output_dir)
create_directory_for_each_family = True if out_prefix else create_dir_for_each_family
if families_id_file:
fam_id_list.read(families_id_file)
fam_dict.read(fam_file, split_values=True, values_separator=",")
protein_dict = SeqIO.index_db("tmp.idx", pep_file, format=pep_format)
for fam_id in fam_id_list if families_id_file else fam_dict:
if fam_id in fam_dict:
if create_directory_for_each_family:
fam_dir = "%s%s/" % (out_dir, fam_id)
FileRoutines.safe_mkdir(fam_dir)
out_file = "%s%s.pep" % (fam_dir, out_prefix
if out_prefix else fam_id)
else:
out_file = "%s/%s.pep" % (out_dir, out_prefix
if out_prefix else fam_id)
SeqIO.write(SequenceRoutines.record_by_id_generator(
protein_dict, fam_dict[fam_id], verbose=True),
out_file,
format=pep_format)
else:
print("%s was not found" % fam_id)
os.remove("tmp.idx")
parser.add_argument("-d",
"--id_file",
action="store",
dest="id_file",
required=True,
help="File with ids of genes to extract")
parser.add_argument("-w",
"--write_comments",
action="store_true",
dest="write_comments",
help="Write comments to output")
args = parser.parse_args()
feature_id_list = IdList()
feature_id_list.read(args.id_file)
with open(args.input, "r") as in_fd:
with open(args.output, "w") as out_fd:
for line in in_fd:
if (line[0] == "#") and args.write_comments:
out_fd.write(line)
continue
description_list = line.split("\t")[9].split(";")
feature_id = description_list[0].split("=")[1]
if feature_id not in feature_id_list:
continue
out_fd.write(line)
while True:
description_list = in_fd.next().split("\t")[9].split(";")
dest="logs_dir",
default="logs",
type=check_path,
help="Directory with logs")
args = parser.parse_args()
save_mkdir(args.output_dir)
save_mkdir(args.logs_dir)
if (not args.alignment) and (not args.tree) and (not args.hmm):
args.all = True
in_fd = sys.stdin if args.input == "stdin" else open(args.input, "r")
family_ids = IdList()
family_ids.read(in_fd)
if args.input != "stdin":
in_fd.close()
absent_alignment_list = IdList()
absent_tree_list = IdList()
absent_hmm_list = IdList()
def download_data(fam_id):
print("Downloading %s family" % fam_id)
ali_log_file = "/dev/null" if not args.store_logs else "%s%s_alignment.log" % (
args.logs_dir, fam_id)
tree_log_file = "/dev/null" if not args.store_logs else "%s%s_tree.log" % (
args.logs_dir, fam_id)
#!/usr/bin/env python
__author__ = 'Sergei F. Kliver'
import argparse
from CustomCollections.GeneralCollections import IdList
from Routines import SequenceRoutines, FileRoutines
parser = argparse.ArgumentParser()
parser.add_argument("-i", "--input", action="store", dest="input", required=True,
type=lambda s: FileRoutines.make_list_of_path_to_files(s.split(",")),
help="Comma-separated list of genbank files/directories with transcript annotations")
parser.add_argument("-o", "--output", action="store", dest="output", required=True,
help="Output file")
parser.add_argument("-d", "--id_file", action="store", dest="id_file",
help="File with id of transcripts to deal with")
args = parser.parse_args()
if args.id_file:
id_list = IdList()
id_list.read(args.id_file)
else:
id_list = None
SequenceRoutines.extract_introns_from_transcripts_from_genbank_files(args.input, args.output,
transcript_id_white_list=id_list)
def get_taxonomy_from_id_file(self, taxa_file, output_file, email, input_type="latin"):
taxa_list = IdList()
taxa_list.read(taxa_file)
self.get_taxonomy(taxa_list, output_file, email, input_type=input_type)
import sys
import argparse
from CustomCollections.GeneralCollections import IdList
parser = argparse.ArgumentParser()
parser.add_argument("-i",
"--fam_file",
action="store",
dest="fam_file",
required=True,
help="File with families")
parser.add_argument("-o",
"--output",
action="store",
dest="output",
default="stdout",
help="File to write ids")
args = parser.parse_args()
out_fd = sys.stdout if args.output == "stdout" else open(args.output, "w")
id_list = IdList()
id_list.read(args.fam_file,
close_after_if_file_object=True,
column_number=1,
id_in_column_separator=",")
id_list.write(args.output, close_after_if_file_object=True)
default="stdout",
help="Prefix of output file")
args = parser.parse_args()
out_fd = sys.stdout if args.output_prefix == "stdout" else open(
"%s_reference_random_genes.ids" % args.output_prefix, "w")
reference_families = SynDict()
reference_families.read(args.reference_fam,
separator="\t",
split_values=True,
values_separator=",")
node_family_ids = IdList()
node_family_ids.read(args.input,
header=True,
column_number=0,
column_separator="\t")
reference_random_genes = SynDict()
for family_id in node_family_ids:
if family_id not in reference_families:
reference_random_genes[family_id] = "."
else:
reference_random_genes[family_id] = choice(
reference_families[family_id])
reference_random_genes.write("%s_reference_random_genes.t" %
args.output_prefix)
for family_id in reference_random_genes:
"--header",
action="store_true",
dest="header",
help="Header is present in id file. Default: False")
parser.add_argument("-f",
"--format",
action="store",
dest="format",
required=True,
help="Format of the file with hits")
parser.add_argument("-o",
"--output_file",
action="store",
dest="output",
default="stdout",
help="Output file")
args = parser.parse_args()
out_fd = sys.stdout if args.output == "stdout" else open(args.output, "w")
id_list = IdList()
id_list = id_list.read(args.id_file, header=args.header)
HMMER3.extract_hits_by_query_ids(id_list,
args.input,
args.output,
fileformat=args.format,
close_after_if_file_object=True)
out_fd.close()
parser.add_argument("-o",
"--output_file",
action="store",
dest="output_file",
help="Output file with extracted_annotations")
parser.add_argument("-d",
"--ids_file",
action="store",
dest="ids_file",
help="File with ids of annotations to extract")
parser.add_argument("-t",
"--annotation_types",
action="store",
dest="annotation_types",
default=["gene"],
type=lambda s: s.split(","),
help="Comma-separated list of annotation types to extract")
args = parser.parse_args()
annotation_ids = IdList()
annotation_ids.read(args.ids_file, comments_prefix="#")
#print args.annotation_types
out_fd = open(args.output_file, "w")
GFF.write(
record_with_extracted_annotations_generator(args.input_gff,
args.annotation_types), out_fd)
out_fd.close()
def get_cds_for_proteins_from_id_file(self, protein_id_file, output_prefix):
pep_ids = IdList()
pep_ids.read(protein_id_file)
self.get_cds_for_proteins(pep_ids, output_prefix)