def create_star_genome_index(fasta_file, out_dir, genome_anno=None, num_workers=1, onematelength=100):
"""
Creating STAR genome index with or without using genome annotation
TODO check whether the fasta and gtf files are uncompressed star works with uncompressed files in this step.
@args fasta_file: reference genome sequence file .fasta format
@type fasta_file: str
@args out_dir: genome index binary file storage place
@type out_dir: str
@args genome_anno: genome annotation file (optional)
@type genome_anno: str
@args num_workers: number of threads to run (default value = 1)
@type num_workers: int
@args onematelength: One Mate Length (default value=100)
@type num_workers: int
"""
try:
subprocess.call(["STAR"], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
except:
exit("Please make sure that the `STAR` binary is in your $PATH")
if not genome_anno:
cli_cmd = 'STAR \
--runMode genomeGenerate \
--genomeDir %s \
--genomeFastaFiles %s \
--runThreadN %d' % (out_dir, fasta_file, num_workers)
else:
## check for the file type
gff_hand = helper.open_file(genome_anno)
for rec in gff_hand:
rec = rec.strip('\n\r')
# skip empty line fasta identifier and commented line
if not rec or rec[0] in ['#', '>']:
continue
# skip the genome sequence
if not re.search('\t', rec):
continue
parts = rec.split('\t')
assert len(parts) >= 8, rec
ftype, tags = GFFParser.attribute_tags(parts[-1])
break
gff_hand.close()
## according to the file type
if ftype:
cli_cmd = 'STAR \
--runMode genomeGenerate \
--genomeDir %s \
--genomeFastaFiles %s \
--runThreadN %d \
--sjdbGTFfile %s \
--sjdbGTFtagExonParentTranscript Parent \
--sjdbOverhang %d' % (out_dir, fasta_file, num_workers, genome_anno, onematelength)
else:
cli_cmd = 'STAR \
--runMode genomeGenerate \
--genomeDir %s \
--genomeFastaFiles %s \
--runThreadN %d \
--sjdbGTFfile %s \
--sjdbGTFfeatureExon exon \
--sjdbOverhang %d' % (out_dir, fasta_file, num_workers, genome_anno, onematelength)
## create downloadpath if doesnot exists
if not os.path.exists(out_dir):
try:
os.makedirs(out_dir)
except OSError:
print "error: cannot create the directory %s." % out_dir
sys.exit(0)
else:## if present any other old index files clean up the folder
for the_file in os.listdir(out_dir):
file_path = os.path.join(out_dir, the_file)
try:
if os.path.isfile(file_path):
os.unlink(file_path)
elif os.path.isdir(file_path):
shutil.rmtree(file_path)
except Exception, e:
print e
def create_star_genome_index(fasta_file,
out_dir,
genome_anno=None,
num_workers=1,
onematelength=100):
"""
Creating STAR genome index with or without using genome annotation
@args fasta_file: reference genome sequence file .fasta format
@type fasta_file: str
@args out_dir: genome index binary file storage place
@type out_dir: str
@args genome_anno: genome annotation file (optional)
@type genome_anno: str
@args num_workers: number of threads to run (default value = 1)
@type num_workers: int
@args onematelength: One Mate Length (default value=100)
@type onematelength: int
"""
try:
subprocess.call(["STAR"],
stdout=subprocess.PIPE,
stderr=subprocess.PIPE)
except:
exit("Please make sure that the `STAR` binary is in your $PATH")
file_prefx, ext = os.path.splitext(fasta_file)
if ext in [".bz2", ".gz", ".lzma"
]: ## checking for the compressed form of the file extension
exit(
"error: STAR - Generating genome indexes - recommended to use the uncompressed FASTA file %s."
% fasta_file)
if not genome_anno:
cli_cmd = 'STAR \
--runMode genomeGenerate \
--genomeDir %s \
--genomeFastaFiles %s \
--runThreadN %d' % (out_dir, fasta_file, num_workers)
else:
file_prefx, ext = os.path.splitext(genome_anno)
if ext in [".bz2", ".gz", ".lzma"]:
exit(
"error: STAR - Generating genome indexes - recommended to use the uncompressed GTF/GFF file %s."
% genome_anno)
## check for the file type
gff_hand = helper.open_file(genome_anno)
for rec in gff_hand:
rec = rec.strip('\n\r')
# skip empty line fasta identifier and commented line
if not rec or rec[0] in ['#', '>']:
continue
# skip the genome sequence
if not re.search('\t', rec):
continue
parts = rec.split('\t')
assert len(parts) >= 8, rec
ftype, tags = GFFParser.attribute_tags(parts[-1])
break
gff_hand.close()
## according to the file type
if ftype:
cli_cmd = 'STAR \
--runMode genomeGenerate \
--genomeDir %s \
--genomeFastaFiles %s \
--runThreadN %d \
--sjdbGTFfile %s \
--sjdbGTFtagExonParentTranscript Parent \
--sjdbOverhang %d' % (out_dir, fasta_file, num_workers,
genome_anno, onematelength)
else:
cli_cmd = 'STAR \
--runMode genomeGenerate \
--genomeDir %s \
--genomeFastaFiles %s \
--runThreadN %d \
--sjdbGTFfile %s \
--sjdbGTFfeatureExon exon \
--sjdbOverhang %d' % (out_dir, fasta_file, num_workers,
genome_anno, onematelength)
## create downloadpath if doesnot exists
if not os.path.exists(out_dir):
try:
os.makedirs(out_dir)
except OSError:
exit("error: cannot create the directory %s." % out_dir)
else: ## if present any other old index files clean up the folder
for the_file in os.listdir(out_dir):
file_path = os.path.join(out_dir, the_file)
try:
if os.path.isfile(file_path):
os.unlink(file_path)
elif os.path.isdir(file_path):
shutil.rmtree(file_path)
except Exception, e:
print(e)
def run_star_alignment(org_db, read_type='PE', max_mates_gap_length=100000, num_cpus=1):
"""
wrapper for running STAR program
@args org_db: a python dictionary with all details about a single organism
@type org_db: defaultdict
@args read_type: library type - paired-end or single-end (default: PE)
@type read_type: str
@args max_mates_gap_length: maximum insert size from the sample (default: 10000)
@type max_mates_gap_length: int
@args num_cpus: number of threads to use for the run (default: 1)
@type num_cpus: int
"""
try:
subprocess.call(["STAR"], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
except:
exit("Please make sure that the `STAR` binary is in your $PATH")
from gfftools import helper, GFFParser
genome_dir = org_db['genome_index_dir']## genome indices and annotation file
gtf_db = org_db['gtf']
if gtf_db != None:
## check for the annotation file type gff or gtf
gff_hand = helper.open_file(gtf_db)
for rec in gff_hand:
rec = rec.strip('\n\r')
# skip empty line fasta identifier and commented line
if not rec or rec[0] in ['#', '>']:
continue
# skip the genome sequence
if not re.search('\t', rec):
continue
parts = rec.split('\t')
assert len(parts) >= 8, rec
ftype, tags = GFFParser.attribute_tags(parts[-1])
break
gff_hand.close()
## library type
if read_type == 'PE':
read_file = "%s/%s %s/%s" % (org_db['fastq_path'], org_db['fastq'][0], org_db['fastq_path'], org_db['fastq'][1])
else:
read_file = "%s/%s" % (org_db['fastq_path'], org_db['fastq'][0])
## getting the command to uncompress the read file
zip_type = {".gz" : "gzip -c", ".bz2" : "bzip2 -d -c"}
file_prefx, ext = os.path.splitext(org_db['fastq'][0])
out_prefix = '%s/%s_' % (org_db['read_map_dir'], org_db['short_name'])
## genomic feature information
max_lenth_intron = org_db['max_intron_len']
## according to the file type
if gtf_db == None:
make_star_run = "STAR \
--genomeDir %s \
--readFilesIn %s \
--readFilesCommand %s \
--outFileNamePrefix %s \
--runThreadN %d \
--outFilterMultimapScoreRange 2 \
--outFilterMultimapNmax 30 \
--outFilterMismatchNmax 4 \
--sjdbScore 1 \
--sjdbOverhang 5 \
--outSAMstrandField intronMotif \
--outFilterIntronMotifs RemoveNoncanonical \
--outSAMtype BAM Unsorted \
--genomeLoad LoadAndRemove" % (genome_dir, read_file,
zip_type[ext], out_prefix, num_cpus)
elif ftype:
make_star_run = "STAR \
--genomeDir %s \
--readFilesIn %s \
--readFilesCommand %s \
--outFileNamePrefix %s \
--runThreadN %d \
--outFilterMultimapScoreRange 2 \
--outFilterMultimapNmax 30 \
--outFilterMismatchNmax 4 \
--alignIntronMax %d \
--sjdbGTFfile %s \
--sjdbGTFtagExonParentTranscript Parent \
--sjdbScore 1 \
--sjdbOverhang 5 \
--outSAMstrandField intronMotif \
--outFilterIntronMotifs RemoveNoncanonical \
--outSAMtype BAM Unsorted \
--genomeLoad LoadAndRemove" % (genome_dir, read_file,
zip_type[ext], out_prefix, num_cpus, max_lenth_intron, gtf_db)
else:
make_star_run = "STAR \
--genomeDir %s \
--readFilesIn %s \
--readFilesCommand %s \
--outFileNamePrefix %s \
--runThreadN %d \
--outFilterMultimapScoreRange 2 \
--outFilterMultimapNmax 30 \
--outFilterMismatchNmax 4 \
--alignIntronMax %d \
--sjdbGTFfile %s \
--sjdbGTFfeatureExon exon \
--sjdbScore 1 \
--sjdbOverhang 5 \
--outSAMstrandField intronMotif \
--outFilterIntronMotifs RemoveNoncanonical \
--outSAMtype BAM Unsorted \
--genomeLoad LoadAndRemove" % (genome_dir, read_file,
zip_type[ext], out_prefix, num_cpus, max_lenth_intron, gtf_db)
sys.stdout.write('\trunning STAR program as: %s \n' % make_star_run)
try:
process = subprocess.Popen(make_star_run, shell=True)
returncode = process.wait()
if returncode !=0:
raise Exception, "Exit status return code = %i" % returncode
sys.stdout.write("STAR run completed. result file stored at %sAligned.out.bam\n" % out_prefix)
except Exception, e:
sys.exit("Error running STAR.\n%s" % str( e ))
def run_star_alignment(org_db, read_type='PE', max_mates_gap_length=100000, num_cpus=1):
"""
wrapper for running STAR program
@args org_db: a python dictionary with all details about a single organism
@type org_db: defaultdict
@args read_type: library type - paired-end or single-end (default: PE)
@type read_type: str
@args max_mates_gap_length: maximum insert size from the sample (default: 10000)
@type max_mates_gap_length: int
@args num_cpus: number of threads to use for the run (default: 1)
@type num_cpus: int
"""
try:
subprocess.call(["STAR"], stdout=subprocess.PIPE, stderr=subprocess.PIPE)
except:
exit("Please make sure that the `STAR` binary is in your $PATH")
from gfftools import helper, GFFParser
genome_dir = org_db['genome_index_dir']## genome indices and annotation file
gtf_db = org_db['gtf']
if gtf_db != None:
## check for the annotation file type gff or gtf
gff_hand = helper.open_file(gtf_db)
for rec in gff_hand:
rec = rec.strip('\n\r')
# skip empty line fasta identifier and commented line
if not rec or rec[0] in ['#', '>']:
continue
# skip the genome sequence
if not re.search('\t', rec):
continue
parts = rec.split('\t')
assert len(parts) >= 8, rec
ftype, tags = GFFParser.attribute_tags(parts[-1])
break
gff_hand.close()
## library type
if read_type == 'PE':
read_file = "%s/%s %s/%s" % (org_db['fastq_path'], org_db['fastq'][0], org_db['fastq_path'], org_db['fastq'][1])
else:
read_file = "%s/%s" % (org_db['fastq_path'], org_db['fastq'][0])
## getting the command to uncompress the read file
zip_type = {".gz" : "gzip -c", ".bz2" : "bzip2 -d -c"}
file_prefx, ext = os.path.splitext(org_db['fastq'][0])
out_prefix = '%s/%s_' % (org_db['read_map_dir'], org_db['short_name'])
## genomic feature information
max_lenth_intron = org_db['max_intron_len']
##sjdbOverhang
mate_len = org_db['mate_length']
## according to the file type
if gtf_db == None:
make_star_run = "STAR \
--genomeDir %s \
--readFilesIn %s \
--readFilesCommand %s \
--outFileNamePrefix %s \
--runThreadN %d \
--outFilterMultimapScoreRange 2 \
--outFilterMultimapNmax 30 \
--outFilterMismatchNmax 3 \
--sjdbScore 1 \
--sjdbOverhang %d \
--outSAMstrandField intronMotif \
--outFilterIntronMotifs RemoveNoncanonical \
--outSAMtype BAM Unsorted \
--genomeLoad LoadAndRemove" % (genome_dir, read_file,
zip_type[ext], out_prefix, num_cpus, mate_len)
elif ftype:
make_star_run = "STAR \
--genomeDir %s \
--readFilesIn %s \
--readFilesCommand %s \
--outFileNamePrefix %s \
--runThreadN %d \
--outFilterMultimapScoreRange 2 \
--outFilterMultimapNmax 30 \
--outFilterMismatchNmax 3 \
--alignIntronMax %d \
--sjdbGTFfile %s \
--sjdbGTFtagExonParentTranscript Parent \
--sjdbScore 1 \
--sjdbOverhang %d \
--outSAMstrandField intronMotif \
--outFilterIntronMotifs RemoveNoncanonical \
--outSAMtype BAM Unsorted \
--genomeLoad LoadAndRemove" % (genome_dir, read_file,
zip_type[ext], out_prefix, num_cpus, max_lenth_intron, gtf_db, mate_len)
else:
make_star_run = "STAR \
--genomeDir %s \
--readFilesIn %s \
--readFilesCommand %s \
--outFileNamePrefix %s \
--runThreadN %d \
--outFilterMultimapScoreRange 2 \
--outFilterMultimapNmax 30 \
--outFilterMismatchNmax 3 \
--alignIntronMax %d \
--sjdbGTFfile %s \
--sjdbGTFfeatureExon exon \
--sjdbScore 1 \
--sjdbOverhang %d \
--outSAMstrandField intronMotif \
--outFilterIntronMotifs RemoveNoncanonical \
--outSAMtype BAM Unsorted \
--genomeLoad LoadAndRemove" % (genome_dir, read_file,
zip_type[ext], out_prefix, num_cpus, max_lenth_intron, gtf_db, mate_len)
sys.stdout.write('\trunning STAR program as: %s \n' % make_star_run)
try:
process = subprocess.Popen(make_star_run, shell=True)
returncode = process.wait()
if returncode !=0:
raise Exception, "Exit status return code = %i" % returncode
sys.stdout.write("STAR run completed. result file stored at %sAligned.out.bam\n" % out_prefix)
except Exception, e:
sys.exit("Error running STAR.\n%s" % str( e ))
