Merge pull request #245 from openvax/fix-reference-name

Fix reference names from CLI

varcode/__init__.py

@@ -24,7 +24,7 @@
     NonsilentCodingMutation,
 )
 
-__version__ = '1.0.2'
+__version__ = '1.0.3'
 
 __all__ = [
     # basic classes

varcode/cli/effects_script.py

@@ -14,7 +14,6 @@
 
 from __future__ import division, absolute_import
 
-import logging
 import logging.config
 import pkg_resources
 import sys

varcode/cli/variant_args.py

@@ -1,4 +1,4 @@
-# Copyright (c) 2016. Mount Sinai School of Medicine
+# Copyright (c) 2016-2019. Mount Sinai School of Medicine
 #
 # Licensed under the Apache License, Version 2.0 (the "License");
 # you may not use this file except in compliance with the License.
@@ -15,8 +15,6 @@
 from __future__ import print_function, division, absolute_import
 from argparse import ArgumentParser
 
-from pyensembl import genome_for_reference_name
-
 from ..vcf import load_vcf
 from ..maf import load_maf
 from ..variant_collection import VariantCollection
@@ -120,21 +118,15 @@ def download_and_install_reference_data(variant_collections):
 def variant_collection_from_args(args, required=True):
     variant_collections = []
 
-    if args.genome:
-        genome = genome_for_reference_name(args.genome)
-    else:
-        # no genome specified, assume it can be inferred from the file(s)
-        # we're loading
-        genome = None
-
     for vcf_path in args.vcf:
-        variant_collections.append(load_vcf(vcf_path, genome=genome))
+        variant_collections.append(
+            load_vcf(vcf_path, genome=args.genome))
 
     for maf_path in args.maf:
         variant_collections.append(load_maf(maf_path))
 
     if args.variant:
-        if not genome:
+        if not args.genome:
             raise ValueError(
                 "--genome must be specified when using --variant")
 
@@ -144,7 +136,7 @@ def variant_collection_from_args(args, required=True):
                 start=position,
                 ref=ref,
                 alt=alt,
-                ensembl=genome)
+                genome=args.genome)
             for (chromosome, position, ref, alt)
             in args.variant
         ]

varcode/maf.py

@@ -79,6 +79,7 @@ def load_maf_dataframe(path, nrows=None, raise_on_error=True, encoding=None):
         low_memory=False,
         skip_blank_lines=True,
         header=0,
+        nrows=nrows,
         encoding=encoding)
 
     if len(df.columns) < n_basic_columns:
@@ -117,7 +118,8 @@ def load_maf(
         sort_key=variant_ascending_position_sort_key,
         distinct=True,
         raise_on_error=True,
-        encoding=None):
+        encoding=None,
+        nrows=None):
     """
     Load reference name and Variant objects from MAF filename.
 
@@ -143,10 +145,17 @@ def load_maf(
 
     encoding : str, optional
         Encoding to use for UTF when reading MAF file.
+
+    nrows : int, optional
+        Limit to number of rows loaded
     """
     # pylint: disable=no-member
     # pylint gets confused by read_csv inside load_maf_dataframe
-    maf_df = load_maf_dataframe(path, raise_on_error=raise_on_error, encoding=encoding)
+    maf_df = load_maf_dataframe(
+        path,
+        nrows=nrows,
+        raise_on_error=raise_on_error,
+        encoding=encoding)
 
     if len(maf_df) == 0 and raise_on_error:
         raise ValueError("Empty MAF file %s" % path)