Gene-list filtering in extract_annot_start.py.

Also have setup.py install the script.

README.md

@@ -5,9 +5,10 @@ This Python package also installs some shell scripts:
 * `sam2indexedbam.sh` (depends on samtools)
 * `bam2bigwig.sh` (depends on bedops, python3, samtools, bedtools, niceload from parallel and bedGraphToBigWig from UCSC-tools)
 
-It also provides a Python script to transfer bigwig data from one region to another:
+It also provides some Python scripts.
 
-* `copypaste_bigwig_regions.py`
+* `copypaste_bigwig_regions.py` to transfer bigwig data from one region to another.
+* `extract_annot_start.py` to extract starting portions of bed records.
 
 
 ## Installing

scripts/extract_annot_start.py

@@ -35,6 +35,11 @@ def main():
         "-b", "--bedfile",
         required=True,
         help="Input bed file.")
+    parser.add_argument(
+        "-g", "--gene_list",
+        help="File containing a list of gene identifiers. "
+        "If provided, only annotations whose 4th column "
+        "matches one of these identifiers will be considered.")
     parser.add_argument(
         "-s", "--start_size",
         type=int,
@@ -48,10 +53,15 @@ def main():
 
     start_size = args.start_size
     keep_short = args.keep_short
+    if args.gene_list:
+        with open(args.gene_list) as gene_list_fh:
+            gene_ids = set(line.strip() for line in gene_list_fh)
     nb_too_short = 0
     with open(args.bedfile) as bedfile:
         for line in bedfile:
             (chrom, start, end, name, score, strand) = line.strip().split("\t")
+            if gene_ids and name not in gene_ids:
+                continue
             if int(end) - int(start) < start_size:
                 nb_too_short += 1
                 if keep_short:

setup.py

@@ -34,7 +34,8 @@ setup(
     packages=find_packages(),
     scripts=[
         "scripts/bam2bigwig.sh", "scripts/sam2indexedbam.sh",
-        "scripts/copypaste_bigwig_regions.py"],
+        "scripts/copypaste_bigwig_regions.py",
+        "scripts/extract_annot_start.py"],
     install_requires=[
         #"libworkflows @ git+https://gitlab+deploy-token-31:isEzpsgbNf2sJMdUDy2g@gitlab.pasteur.fr/bli/libworkflows.git@744dd79b579577cb6e131653260d7990946be3ad#egg=libworkflows-0.1",
         #"libworkflows @ git+https://gitlab+deploy-token-31:isEzpsgbNf2sJMdUDy2g@gitlab.pasteur.fr/bli/libworkflows.git#egg=libworkflows-0.1",