From a6434a924bd52a69a6d99e507e0bb01bc7747de1 Mon Sep 17 00:00:00 2001
From: Martijn Vermaat <martijn@vermaat.name>
Date: Tue, 16 Aug 2011 10:56:39 +0000
Subject: [PATCH] Fix: don't remove flanking splice sites on deletion of exons
on a transcript reference.
git-svn-id: https://humgenprojects.lumc.nl/svn/mutalyzer/branches/refactor-mutalyzer-branch@322 eb6bd6ab-9ccd-42b9-aceb-e2899b4a52f1
---
README | 2 +-
mutalyzer/variantchecker.py | 29 ++++++-
...st_mutalyzer.py => test_variantchecker.py} | 79 ++++++++++++++-----
tests/test_website.py | 4 +-
4 files changed, 90 insertions(+), 24 deletions(-)
rename tests/{test_mutalyzer.py => test_variantchecker.py} (78%)
diff --git a/README b/README
index 616fbac1..0774f980 100644
--- a/README
+++ b/README
@@ -73,7 +73,7 @@ Todo list:
- Check for os.path.join vulnerabilities.
- Use web.config.debug=False on production server and perhaps put this in
the configuration file.
-- Add database indices to extras/post-install.sh script.
+- Solution for database schema migration on version updates.
Code style guide:
- Follow PEP 8 (code) and PEP 257 (docstrings).
diff --git a/mutalyzer/variantchecker.py b/mutalyzer/variantchecker.py
index 99896625..f717c3ca 100644
--- a/mutalyzer/variantchecker.py
+++ b/mutalyzer/variantchecker.py
@@ -978,9 +978,32 @@ def process_raw_variant(mutator, variant, record, transcript, output):
if transcript and variant.MutationType == 'del':
removed_sites = []
for acceptor, donor in util.grouper(transcript.CM.RNA):
- if first <= acceptor <= last + 1:
+
+ # If we have introns, we match splice sites in a fuzzy way. This
+ # Means that in the case of
+ #
+ # a b
+ # ===========------------=============
+ #
+ # with splice sites a and b, a deletion a+1_b-1 of the entire
+ # intron gets treated as a deletion of both splice sites.
+ #
+ # We don't want this behaviour on e.g. RNA, where we only have
+ # exons. In the case of
+ #
+ # a b c d
+ # ========== ============= ===========
+ #
+ # with splice sites a b c d, a deletion b_c of the middle exon
+ # should only remove splice sites b and c, not a and d.
+ if record.record.molType == 'g':
+ fuzzy = 1
+ else:
+ fuzzy = 0
+
+ if first <= acceptor <= last + fuzzy:
removed_sites.append(acceptor)
- if first - 1 <= donor <= last:
+ if first - fuzzy <= donor <= last:
removed_sites.append(donor)
if len(removed_sites) and not len(removed_sites) % 2:
@@ -1004,6 +1027,8 @@ def process_raw_variant(mutator, variant, record, transcript, output):
output.addMessage(__file__, 1, 'IDELSPLICE',
'Removed %i splice sites from selected ' \
'transcript.' % len(removed_sites))
+ # This is primarily for use in unittests.
+ output.addOutput('removedSpliceSites', len(removed_sites))
# If splice_abort is set, this basically means WOVERSPLICE was called and
# IDELSPLICE was not called.
diff --git a/tests/test_mutalyzer.py b/tests/test_variantchecker.py
similarity index 78%
rename from tests/test_mutalyzer.py
rename to tests/test_variantchecker.py
index ddbaf0c3..1fba607f 100644
--- a/tests/test_mutalyzer.py
+++ b/tests/test_variantchecker.py
@@ -1,5 +1,5 @@
"""
-Tests for the Mutalyzer module.
+Tests for the variantchecker module.
"""
@@ -16,14 +16,13 @@ from mutalyzer.output import Output
from mutalyzer.variantchecker import check_variant
-class TestMutalyzer():
+class TestVariantchecker():
"""
- Test the Mutalyzer module.
+ Test the variantchecker module.
"""
-
def setUp(self):
"""
- Initialize test Mutalyzer module.
+ Initialize test variantchecker module.
"""
self.config = Config()
self.output = Output(__file__, self.config.Output)
@@ -42,7 +41,7 @@ class TestMutalyzer():
"""
check_variant('NM_003002.2:c.274del', self.config, self.output)
wroll = self.output.getMessagesWithErrorCode('WROLLFORWARD')
- assert len(wroll) == 0
+ assert_equal(len(wroll), 0)
def test_no_roll_splice(self):
"""
@@ -52,7 +51,7 @@ class TestMutalyzer():
wrollback = self.output.getMessagesWithErrorCode('IROLLBACK')
assert len(wrollback) > 0
wroll = self.output.getMessagesWithErrorCode('WROLLFORWARD')
- assert len(wroll) == 0
+ assert_equal(len(wroll), 0)
def test_partial_roll_splice(self):
"""
@@ -104,7 +103,7 @@ class TestMutalyzer():
check_variant('AL449423.14:g.65470_65471insTAC', self.config, self.output)
assert 'AL449423.14(CDKN2A_v001):c.99_100insTAG' in self.output.getOutput('descriptions')
assert_equal ('AL449423.14:g.65471_65472insACT', self.output.getIndexedOutput('genomicDescription', 0, ''))
- assert len(self.output.getMessagesWithErrorCode('WROLLFORWARD')) == 1
+ assert_equal(len(self.output.getMessagesWithErrorCode('WROLLFORWARD')), 1)
def test_roll_reverse_ins(self):
"""
@@ -114,7 +113,7 @@ class TestMutalyzer():
check_variant('AL449423.14:g.65471_65472insACT', self.config, self.output)
assert 'AL449423.14(CDKN2A_v001):c.99_100insTAG' in self.output.getOutput('descriptions')
assert_equal ('AL449423.14:g.65471_65472insACT', self.output.getIndexedOutput('genomicDescription', 0, ''))
- assert len(self.output.getMessagesWithErrorCode('WROLLFORWARD')) == 0
+ assert_equal(len(self.output.getMessagesWithErrorCode('WROLLFORWARD')), 0)
def test_roll_message_forward(self):
"""
@@ -122,8 +121,8 @@ class TestMutalyzer():
strand (forward).
"""
check_variant('AL449423.14:g.65470_65471insTAC', self.config, self.output)
- assert len(self.output.getMessagesWithErrorCode('WROLLFORWARD')) == 1
- assert len(self.output.getMessagesWithErrorCode('WROLLREVERSE')) == 0
+ assert_equal(len(self.output.getMessagesWithErrorCode('WROLLFORWARD')), 1)
+ assert_equal(len(self.output.getMessagesWithErrorCode('WROLLREVERSE')), 0)
def test_roll_message_reverse(self):
"""
@@ -131,8 +130,8 @@ class TestMutalyzer():
strand (reverse).
"""
check_variant('AL449423.14(CDKN2A_v001):c.98_99insGTA', self.config, self.output)
- assert len(self.output.getMessagesWithErrorCode('WROLLFORWARD')) == 0
- assert len(self.output.getMessagesWithErrorCode('WROLLREVERSE')) == 1
+ assert_equal(len(self.output.getMessagesWithErrorCode('WROLLFORWARD')), 0)
+ assert_equal(len(self.output.getMessagesWithErrorCode('WROLLREVERSE')), 1)
def test_ins_cds_start(self):
"""
@@ -157,7 +156,7 @@ class TestMutalyzer():
check_variant('NG_012772.1(BRCA2_v001):c.632-5_670del',
self.config, self.output)
assert len(self.output.getMessagesWithErrorCode('WOVERSPLICE')) > 0
- assert len(self.output.getMessagesWithErrorCode('IDELSPLICE')) == 0
+ assert_equal(self.output.getOutput('removedSpliceSites'), [])
# Todo: For now, the following is how to check if no protein
# prediction is done.
assert not self.output.getOutput('newprotein')
@@ -169,7 +168,19 @@ class TestMutalyzer():
check_variant('NG_012772.1(BRCA2_v001):c.632-5_681+7del',
self.config, self.output)
assert len(self.output.getMessagesWithErrorCode('WOVERSPLICE')) > 0
- assert len(self.output.getMessagesWithErrorCode('IDELSPLICE')) > 0
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
+ # Todo: For now, the following is how to check if protein
+ # prediction is done.
+ assert self.output.getOutput('newprotein')
+
+ def test_del_exon_exact(self):
+ """
+ Deletion of exactly an exon should be possible.
+ """
+ check_variant('NG_012772.1(BRCA2_v001):c.632_681del',
+ self.config, self.output)
+ assert_equal(len(self.output.getMessagesWithErrorCode('WOVERSPLICE')), 0)
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
# Todo: For now, the following is how to check if protein
# prediction is done.
assert self.output.getOutput('newprotein')
@@ -186,7 +197,7 @@ class TestMutalyzer():
check_variant('NG_012772.1(BRCA2_v001):c.68-7_316+7del',
self.config, self.output)
assert len(self.output.getMessagesWithErrorCode('WOVERSPLICE')) > 0
- assert len(self.output.getMessagesWithErrorCode('IDELSPLICE')) > 0
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
# Todo: For now, the following is how to check if protein
# prediction is done.
assert self.output.getOutput('newprotein')
@@ -199,7 +210,7 @@ class TestMutalyzer():
check_variant('NG_012772.1(BRCA2_v001):c.632-5_793+7del',
self.config, self.output)
assert len(self.output.getMessagesWithErrorCode('WOVERSPLICE')) > 0
- assert len(self.output.getMessagesWithErrorCode('IDELSPLICE')) > 0
+ assert_equal(self.output.getOutput('removedSpliceSites'), [4])
# Todo: For now, the following is how to check if protein
# prediction is done.
assert self.output.getOutput('newprotein')
@@ -212,11 +223,27 @@ class TestMutalyzer():
check_variant('NG_012772.1(BRCA2_v001):c.622_674del',
self.config, self.output)
assert len(self.output.getMessagesWithErrorCode('WOVERSPLICE')) > 0
- assert len(self.output.getMessagesWithErrorCode('IDELSPLICE')) > 0
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
# Todo: For now, the following is how to check if protein
# prediction is done.
assert self.output.getOutput('newprotein')
+ def test_del_intron_exact(self):
+ """
+ Deletion of exactly an intron should be possible (fusion of flanking
+ exons).
+ """
+ check_variant('NG_012772.1(BRCA2_v001):c.681+1_682-1del',
+ self.config, self.output)
+ assert_equal(self.output.getMessagesWithErrorCode('WOVERSPLICE'), [])
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
+ # Note: The protein prediction is done, but 'newprotein' is not set
+ # because we have no change. So to check if the prediction is done, we
+ # check if 'oldprotein' is set and to check if the prediction is
+ # correct, we check if 'newprotein' is not set.
+ assert self.output.getOutput('oldprotein')
+ assert not self.output.getOutput('newprotein')
+
def test_del_intron_in_frame(self):
"""
Deletion of an entire intron should be possible (fusion of remaining
@@ -225,7 +252,7 @@ class TestMutalyzer():
check_variant('NG_012772.1(BRCA2_v001):c.622_672del',
self.config, self.output)
assert len(self.output.getMessagesWithErrorCode('WOVERSPLICE')) > 0
- assert len(self.output.getMessagesWithErrorCode('IDELSPLICE')) > 0
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
# Todo: For now, the following is how to check if protein
# prediction is done.
assert self.output.getOutput('newprotein')
@@ -296,3 +323,17 @@ class TestMutalyzer():
in self.output.getOutput('descriptions')
# Todo: .c notation should still be c.632-?_681+?del, but what about
# other transcripts?
+
+ def test_del_exon_transcript_reference(self):
+ """
+ Deletion of entire exon on a transcript reference should remove the
+ expected splice sites (only that of the deleted exon), and not those
+ of the flanking exons (as would happen using the mechanism for genomic
+ references).
+ """
+ check_variant('NM_018723.3:c.758_890del', self.config, self.output)
+ assert_equal(len(self.output.getMessagesWithErrorCode('WOVERSPLICE')), 0)
+ assert_equal(self.output.getOutput('removedSpliceSites'), [2])
+ # Todo: For now, the following is how to check if protein
+ # prediction is done.
+ assert self.output.getOutput('newprotein')
diff --git a/tests/test_website.py b/tests/test_website.py
index e740edd4..de31f2c7 100644
--- a/tests/test_website.py
+++ b/tests/test_website.py
@@ -520,8 +520,8 @@ facilisi."""
r.mustcontain('0 Errors',
'0 Warnings',
'Raw variant 1: substitution at 7055')
- assert r.body.find('go to bottom') == -1
- assert r.body.find('<input') == -1
+ assert_equal(r.body.find('go to bottom'), -1)
+ assert_equal(r.body.find('<input'), -1)
def test_variantinfo_g2c(self):
"""
--
GitLab