Support for LRG 1.9

- Switch to new LRG location (fix for #447).
- Extract only the gene name from the `updatable` section.
- Fix for LRG transcripts with no coding region.
- Adapted LRG examples on the website.
- Extract the annotation set based on attribute type.
- More informative error message (part of #135).
- Makes #338 obsolete.

mutalyzer/config/default_settings.py

@@ -65,8 +65,7 @@ OUTPUT_LEVEL = 1
 LOG_TIME_FORMAT = "%Y-%m-%d %H:%M:%S"
 
 # Prefix URL from where LRG files are fetched.
-#LRG_PREFIX_URL = 'ftp://ftp.ebi.ac.uk/pub/databases/lrgex/'
-LRG_PREFIX_URL = 'ftp://ftp.ebi.ac.uk/pub/databases/lrgex/SCHEMA_1_7_ARCHIVE/'
+LRG_PREFIX_URL = 'ftp://ftp.ebi.ac.uk/pub/databases/lrgex/'
 
 # Allow for this fraction of errors in batch jobs.
 BATCH_JOBS_ERROR_THRESHOLD = 0.05

mutalyzer/variantchecker.py

@@ -1551,10 +1551,20 @@ def process_variant(mutator, description, record, output):
                     # Todo: Incorrect error message, it might also be that
                     # there are no transcripts at all (e.g. N4BP2L1 on
                     # NG_012772.1).
-                    output.addMessage(__file__, 4, "ENOTRANSCRIPT",
-                        "Multiple transcripts found for gene %s. Please " \
-                        "choose from: %s" % (gene.name,
-                            ", ".join(gene.listLoci())))
+                    transcripts_no = len(gene.listLoci())
+                    if transcripts_no == 1:
+                        message = "Transcript t%s not found. The reference " \
+                                  "sequence contains only transcript t%s." \
+                                  %(transcript_id, gene.listLoci()[0])
+                    elif transcripts_no > 1:
+                        message = "Transcript t%s not found. The reference " \
+                                  "sequence contains for gene %s the " \
+                                  "following %s transcripts: %s."\
+                                  %(transcript_id, gene.name,
+                                    len(gene.listLoci()),
+                                    ", ".join(("t%s" %c)
+                                              for c in gene.listLoci()))
+                    output.addMessage(__file__, 4, "ENOTRANSCRIPT", message)
             else:
                 # No transcript id given.
                 if len(gene.transcriptList) == 1:

mutalyzer/website/templates/homepage.html

@@ -21,7 +21,7 @@ Variation Society</a>.
                 <p>Examples:
                   <code class="example-input" data-for="description">AB026906.1:c.40_42del</code>,
                   <code class="example-input" data-for="description">NG_012337.1(SDHD_v001):c.274G&gt;T</code>,
-                  <code class="example-input" data-for="description">LRG_9t1:c.159dup</code>
+                  <code class="example-input" data-for="description">LRG_24t1:c.159dup</code>
                 </p>
 
                 <!-- <a href="{{ url_for('.name_checker') }}" class="btn btn-default btn-small btn-primary">Try

mutalyzer/website/templates/name-checker.html

@@ -55,7 +55,7 @@
       <p>Examples:
         <code class="example-input" data-for="description">AB026906.1:c.40_42del</code>,
         <code class="example-input" data-for="description">NG_012337.1(SDHD_v001):c.274G&gt;T</code>,
-        <code class="example-input" data-for="description">LRG_9t1:c.159dup</code>
+        <code class="example-input" data-for="description">LRG_24t1:c.159dup</code>
       </p>
     </div>
 

mutalyzer/website/templates/position-converter.html

@@ -33,7 +33,7 @@ normalize it to HGVS. Use the <a href="{{ url_for('.name_checker') }}">Name Chec
       <code class="example-input"
             data-for="description">NM_003002.3:c.274G&gt;T</code>,
       <code class="example-input"
-            data-for="description">LRG_9t1:c.274G&gt;T</code>,
+            data-for="description">LRG_11t1:c.274G&gt;T</code>,
       <code class="example-input"
             data-for="description">chr11:g.111959693G&gt;T</code>,
       <code class="example-input"

mutalyzer/website/templates/syntax-checker.html

@@ -20,7 +20,7 @@ Sequence Variant Nomenclature">HGVS</a> format.
     <p>Examples:
       <code class="example-input" data-for="description">AB026906.1:c.40_42del</code>,
       <code class="example-input" data-for="description">NG_012337.1(SDHD_v001):c.274G&gt;T</code>,
-      <code class="example-input" data-for="description">LRG_9t1:c.159dup</code>
+      <code class="example-input" data-for="description">LRG_24t1:c.159dup</code>
     </p>
   </div>
   <div class="form-group button-group">

tests/data/references.yml

@@ -147,8 +147,14 @@ L41870.1:
   checksum: 91b1e539a053f731f95d230a06710897
   filename: L41870.1.gb.bz2
 LRG_1:
-  checksum: 5b8f5a39fcd9e3005688eddffd482746
+  checksum: 8dc3c2fcc80b3216319d8b218e5093ab
   filename: LRG_1.xml.bz2
+LRG_24:
+  checksum: 551d043bd217004059525999a134ef3b
+  filename: LRG_24.xml.bz2
+LRG_163:
+  checksum: fd4a3161a23ba1ed23c112195f14617d
+  filename: LRG_163.xml.bz2
 MARK1:
   accession: UD_139015213982
   checksum: 0d63a8fe5beddeb793940f6ae194b985

tests/test_parsers_lrg.py

+"""
+Tests for the mutalyzer.parsers.lrg module.
+"""
+
+
+from __future__ import unicode_literals
+
+import os
+import bz2
+
+from mutalyzer.parsers.lrg import create_record
+
+from fixtures import with_references
+
+
+@with_references('LRG_1')
+def test_lrg_basic(settings, references):
+    """
+    """
+    accession = references[0].accession
+    filename = os.path.join(settings.CACHE_DIR, '%s.xml.bz2' % accession)
+    file_handle = bz2.BZ2File(filename, 'r')
+    record = create_record(file_handle.read())
+    file_handle.close()
+
+    assert [g.name for g in record.geneList] == ['COL1A1']
+    assert record.geneList[0].transcriptList[0].name == '1'
+    assert len(record.geneList[0].transcriptList) == 1
+    assert record.geneList[0].transcriptList[0].CDS is not None
+    assert record.organism == 'Homo sapiens'
+
+
+@with_references('LRG_24')
+def test_lrg_multiple_transcripts(settings, references):
+    """
+    """
+    accession = references[0].accession
+    filename = os.path.join(settings.CACHE_DIR, '%s.xml.bz2' % accession)
+    file_handle = bz2.BZ2File(filename, 'r')
+    record = create_record(file_handle.read())
+    file_handle.close()
+
+    assert len(record.geneList[0].transcriptList) == 2
+
+
+@with_references('LRG_163')
+def test_lrg_no_coding_sequence(settings, references):
+    """
+    """
+    accession = references[0].accession
+    filename = os.path.join(settings.CACHE_DIR, '%s.xml.bz2' % accession)
+    file_handle = bz2.BZ2File(filename, 'r')
+    record = create_record(file_handle.read())
+    file_handle.close()
+
+    assert len(record.geneList[0].transcriptList) == 1
+    assert record.geneList[0].transcriptList[0].CDS is None