def exon_graph(depth, refseq, prefix, outdir=os.getcwd(), trans=None, genes=None, threads=0): beds = HGVS(refseq, trans=trans, genes=genes) messages = beds.get_exons() threads = cpu_count() if not int(threads) else int(threads) pool = multiprocessing.Pool(processes=threads) for transcript, trans_messages in messages.iteritems(): pool.apply_async(graph, (depth, trans_messages, outdir, prefix)) pool.close() pool.join()
def __init__(self, reference, trans=None, genes=None):
self.refdb = os.path.join(database_dir, "transdb",
"ncbi_anno_rel104.dbref.db")
reference = os.path.abspath(reference)
self.HGVS = HGVS(self.refdb, reference, trans, genes)
self._dbhandles = list()
dbtitle = defaultdict(float)
for f in glob(os.path.join(database_dir, "*", "*.bedanno.config")):
dbs = AnnoVar(os.path.abspath(f))
t = "\t".join([str(i) for i in sorted(dbs.search_value.values())])
dbtitle[t] = dbs.order
self._dbhandles.append(dbs)
self.dbtitle = "\t".join(
[j for j in sorted(dbtitle.keys(), key=lambda x: dbtitle[x])])
def __init__(self, **kwargs):
self.outdir = os.path.abspath(kwargs["indir"])
self.win_len = int(kwargs["correct_win_len"]) or 30
self.shift_len = int(kwargs["correct_shift_len"]) or 25
self.contral_wins = int(100.0 / self.shift_len + 0.5) + 1
chroms = str(kwargs["chrom"]).split(",") if kwargs["chrom"] else None
samples = str(kwargs["sample"]).split(",") if kwargs["sample"] else None
all_samples = set()
contigs = list()
self.cnvdata = defaultdict(dict)
self.sample_win_data = defaultdict(dict)
for cnv_data in glob(os.path.join(self.outdir, "chr*.cnv.args")):
chrom = ".".join(os.path.basename(cnv_data).split(".")[0:-2])
if chroms is not None and chrom not in chroms:
continue
cnvdata = SaveLoad(cnv_data)
cnvdata = cnvdata.load()
contigs.append(chrom)
for sample in cnvdata.keys():
dep_f = os.path.join(self.outdir, sample, "%s.W%iS%i.fixdep.gz" % (chrom, self.win_len, self.shift_len))
if os.path.isfile(dep_f) and os.path.isfile(dep_f + '.tbi'):
self.sample_win_data[chrom][sample] = dep_f
if samples is not None and sample not in samples:
continue
all_samples.add(sample)
self.cnvdata[sample][chrom] = cnvdata[sample]
self.samples = sorted(all_samples)
self.contigs = sorted(contigs, key=lambda x: _chrom_valued(x))
databases = os.path.abspath(kwargs["dbdir"])
t_db = os.path.abspath(kwargs["transdb"]) if "transdb" in kwargs else os.path.join(databases, "transdb",
"ncbi_anno_rel104.dbref.db")
for db in glob(os.path.join(databases, "*", "*.cnvdb.config")):
db = os.path.abspath(db)
dbname = os.path.basename(os.path.dirname(db))
_AnnotationDB[dbname].add(db)
self.reference = os.path.abspath(kwargs["reference"]) if kwargs["reference"] else \
os.path.join(databases, 'aln_db/hg19/hg19_chM_male_mask.fa')
self.DBAnno = CNVAnnotation(self.reference, _AnnotationDB)
self.HGVS = HGVS(t_db)
