From 73f457b95ded278e12321afacc76be1ca650dd46 Mon Sep 17 00:00:00 2001
From: Eric Talevich <eric.talevich@gmail.com>
Date: Fri, 1 May 2015 16:41:03 -0700
Subject: [PATCH] doc: explain the new -x option in genome2access.py

---
 doc/scripts.rst          | 37 ++++++++++++++++++++++++++++---------
 scripts/genome2access.py |  2 +-
 2 files changed, 29 insertions(+), 10 deletions(-)

diff --git a/doc/scripts.rst b/doc/scripts.rst
index 2e79962a..80644b4a 100644
--- a/doc/scripts.rst
+++ b/doc/scripts.rst
@@ -1,17 +1,36 @@
 Additional scripts
 ==================
 
-refFlat2bed.py
-    Generate a BED file of the genes or exons in the reference genome given in
-    UCSC refFlat.txt format.
-    This script can be used in case the original BED file of targeted intervals
-    is unavailable.  Subsequent steps of the pipeline will remove probes that
-    did not receive sufficient coverage, including those exons or genes that
-    were not targeted by the sequencing library.  However, better results are
-    expected from CNVkit if the true targeted intervals can be provided.
-
 genome2access.py:
     Calculate the sequence-accessible coordinates in chromosomes from the given
     reference genome, treating long spans of 'N' characters as the inaccessible
     regions.
 
+    CNVkit will compute "antitarget" bins only within the accessible genomic
+    regions specified in the "access" file produced by this script. If there are
+    many small excluded/inaccessible regions in the genome, then small,
+    less-reliable antitarget bins would be squeezed into the remaining
+    accessible regions.  The ``-s`` option tells the script to ignore short
+    regions that would otherwise be excluded as inaccessible, allowing larger
+    antitarget bins to overlap them.
+
+    Additional regions to exclude can also be given with the ``-x`` option. This
+    option can be used more than once to exclude several BED files listing
+    different sets of regions. For example, "excludable" regions of poor
+    mappability have been precalculated by others and are available from the
+    `UCSC FTP Server <ftp://hgdownload.soe.ucsc.edu/goldenPath/>`_
+    (see `here for hg19
+    <ftp://hgdownload.soe.ucsc.edu/goldenPath/hg19/encodeDCC/wgEncodeMapability/>`_).
+
+
+refFlat2bed.py
+    Generate a BED file of the genes or exons in the reference genome given in
+    UCSC refFlat.txt format.  (Download the input file from `UCSC Genome
+    Bioinformatics <http://hgdownload.soe.ucsc.edu/downloads.html>`_).
+
+    This script can be used in case the original BED file of targeted intervals
+    is unavailable. Subsequent steps of the pipeline will remove probes that
+    did not receive sufficient coverage, including those exons or genes that
+    were not targeted by the sequencing library.  However, CNVkit will give much
+    better results if the true targeted intervals can be provided.
+
diff --git a/scripts/genome2access.py b/scripts/genome2access.py
index e9c439d1..07f477e9 100755
--- a/scripts/genome2access.py
+++ b/scripts/genome2access.py
@@ -174,7 +174,7 @@ def next_or_inf(iterable):
     AP = argparse.ArgumentParser(description=__doc__)
     AP.add_argument("fa_fname",
                     help="Genome FASTA file name")
-    AP.add_argument("-s", "--min-gap-size", type=int, default=100,
+    AP.add_argument("-s", "--min-gap-size", type=int, default=5000,
                     help="""Minimum gap size between accessible sequence
                     regions.  Regions separated by less than this distance will
                     be joined together. [Default: %(default)s]""")