genmod 3.10__tar.gz → 3.10.2__tar.gz

{genmod-3.10 → genmod-3.10.2}/CHANGELOG.md

@@ -8,6 +8,19 @@ Please add a new candidate release at the top after changing the latest one. Fee
 
 Try to use the following format:
 
+## [3.10.2]
+### Fixed 
+- Add scoring normalisation for flag lookup mode ([#177](https://github.com/Clinical-Genomics/genmod/pull/177))
+- Fix crash on test for missing annotation key for phased mode ([#178](https://github.com/Clinical-Genomics/genmod/pull/178))
+- Bugfixes related to multiprocessing stability and error handling ([#183](https://github.com/Clinical-Genomics/genmod/pull/183) and [#186](https://github.com/Clinical-Genomics/genmod/pull/186))
+- Removed unused dependency ([#182](https://github.com/Clinical-Genomics/genmod/pull/182))
+
+## [3.10.1]
+### Fixed
+- Documentation formatting for more modern mkdocs (pages to nav) ([#168](https://github.com/Clinical-Genomics/genmod/pull/168))
+- Stable sort in bioconda (BusyBox sort) not recognizing long options ([#170](https://github.com/Clinical-Genomics/genmod/pull/170))
+- Unstable `Annotation=` order in `genmod annotate --regions` ([#173](https://github.com/Clinical-Genomics/genmod/pull/173))
+
 ## [3.10]
 ### Changed
 - Use `uv` with `hatchling` to build and publish ([#104](https://github.com/Clinical-Genomics/genmod/issues/143))

{genmod-3.10 → genmod-3.10.2}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: genmod
-Version: 3.10
+Version: 3.10.2
 Summary: Annotate genetic inheritance models in Variant Call Format (VCF) files
 Project-URL: Repository, http://github.com/Clinical-Genomics/genmod
 Project-URL: Changelog, https://github.com/Clinical-Genomics/genmod/blob/main/CHANGELOG.md
@@ -29,7 +29,6 @@ Requires-Dist: intervaltree>=3.1.0
 Requires-Dist: ped-parser
 Requires-Dist: pytabix
 Requires-Dist: six
-Requires-Dist: vcftoolbox
 Description-Content-Type: text/markdown
 
 # GENMOD #

{genmod-3.10 → genmod-3.10.2}/docs/commands/score-compounds.md

@@ -3,7 +3,7 @@
 This module performs ranking of compound variants.
 
 > [!WARNING]
-> Ranking of compound variants is only done for the the first family in the VCF.
+> Ranking of compound variants is only done for the first family in the VCF.
 
 During the ranking of these compounds the rank score might be modified in place.
 See `genmod/score_variants/compound_scorer.py:L248`.

{genmod-3.10 → genmod-3.10.2}/docs/index.md

@@ -2,7 +2,7 @@
 
 <p align="center">
   <a href="https://github.com/Clinical-Genomics/genmod">
-    <img src="https://github.com/Clinical-Genomics/genmod/raw/master/artwork/tree_man.JPG"/>
+    <img src="https://github.com/Clinical-Genomics/genmod/blob/main/artwork/tree_man.JPG"/>
   </a>
 </p>
 
@@ -10,7 +10,6 @@ Annotating and analyzing primarily genetic models in the [VCF](http://samtools.g
 
 ## Overview ##
 
-
 The main function is to annotate **accurate patterns of inheritance** for each 
 variant in a vcf file including one or several families of arbitrary size. 
 

{genmod-3.10 → genmod-3.10.2}/genmod/annotate_models/variant_annotator.py

@@ -130,7 +130,7 @@ class VariantAnnotator(Process):
 
                     variant["compound_candidate"] = False
 
-                    if variant["annotation"]:
+                    if variant.get("annotation"):
                         variant["compound_candidate"] = True
                         self.logger.debug("Set compound_candidate to True")
 

{genmod-3.10 → genmod-3.10.2}/genmod/annotate_variants/annotate.py

@@ -37,8 +37,9 @@ def annotate_variant(variant, annotation_arguments):
     regions = None
     if "region_trees" in annotation_arguments:
         regions = get_region(chrom, start, end, annotation_arguments["region_trees"])
+        sorted_regions = sorted(regions)
         if regions:
-            info.append("Annotation={0}".format(",".join(regions)))
+            info.append("Annotation={0}".format(",".join(sorted_regions)))
 
     if "exac" in annotation_arguments:
         reader = annotation_arguments["exac"]

{genmod-3.10 → genmod-3.10.2}/genmod/commands/annotate_models.py

@@ -317,8 +317,13 @@ def models(
         logger.info("Seting up the variant printer")
         if len(model_checkers) == 1:
             print_headers(head=head, outfile=outfile, silent=silent)
+            if hasattr(outfile, "name"):
+                outfile_name = outfile.name
+                outfile.close()
+            else:
+                outfile_name = outfile
             variant_printer = VariantPrinter(
-                task_queue=results, head=head, mode="normal", outfile=outfile
+                task_queue=results, head=head, mode="normal", outfile=outfile_name
             )
         else:
             # We use a temp file to store the processed variants

{genmod-3.10 → genmod-3.10.2}/genmod/commands/score_compounds.py

@@ -17,8 +17,9 @@ import os
 import sys
 from codecs import open
 from datetime import datetime
-from multiprocessing import JoinableQueue, Manager, cpu_count, util
+from multiprocessing import JoinableQueue, Manager, cpu_count, log_to_stderr, util
 from tempfile import NamedTemporaryFile
+from time import sleep
 
 import click
 
@@ -29,7 +30,8 @@ from genmod.vcf_tools import HeaderParser, add_metadata, print_headers, print_va
 
 from .utils import get_file_handle, outfile, processes, silent, temp_dir, variant_file
 
-logger = logging.getLogger(__name__)
+logger = log_to_stderr(logging.INFO)
+logging.basicConfig(stream=sys.stderr, force=True)
 util.abstract_sockets_supported = False
 
 
@@ -160,6 +162,15 @@ def compound(
         for i in range(num_scorers):
             variant_queue.put(None)
 
+        # Before joining on variant_queue, check whether workers have completed
+        # or failed, to avoid main process deadlock on never-decreasing queue semaphore.
+        while any([worker.is_alive() for worker in compound_scorers]):
+            sleep(1)  # Don't churn CPU
+            for worker in compound_scorers:
+                if not worker.is_alive() and worker.exitcode != 0:
+                    raise RuntimeError(f"Worker {worker} failed")
+                logger.debug(f"Worker {worker} alive")
+
         variant_queue.join()
         results.put(None)
         variant_printer.join()
@@ -172,7 +183,7 @@ def compound(
             for line in f:
                 print_variant(variant_line=line, outfile=outfile, mode="modified", silent=silent)
     except Exception as e:
-        logger.warning(e)
+        logger.error(e)
         for worker in compound_scorers:
             worker.terminate()
         variant_printer.terminate()

{genmod-3.10 → genmod-3.10.2}/genmod/score_variants/compound_scorer.py

@@ -14,7 +14,8 @@ Copyright (c) 2013 __MyCompanyName__. All rights reserved.
 from __future__ import division, print_function
 
 import logging
-from multiprocessing import Process
+import traceback
+from multiprocessing import Process, log_to_stderr
 from typing import Dict, List, Tuple, Union
 
 from genmod.score_variants.cap_rank_score_to_min_bound import cap_rank_score_to_min_bound
@@ -26,7 +27,7 @@ from genmod.score_variants.score_variant import (
 )
 from genmod.vcf_tools import add_vcf_info, replace_vcf_info
 
-logger = logging.getLogger(__name__)
+logger = log_to_stderr(level=logging.INFO)
 
 
 def get_rank_score(
@@ -169,7 +170,7 @@ class CompoundScorer(Process):
             )
         return variant_rankscore_normalization_bounds
 
-    def run(self):
+    def _run(self):
         """Run the consuming"""
         logger.info("%s: Starting!" % self.proc_name)
         # Check if there are any batches in the queue
@@ -349,3 +350,11 @@ class CompoundScorer(Process):
             self.task_queue.task_done()
 
         return
+
+    def run(self, *args, **kwargs):
+        # Wrapper for catching errors in main method
+        try:
+            return self._run(*args, **kwargs)
+        except Exception as e:
+            logger.fatal(f"{e}:{traceback.format_exc()}")
+            raise e

{genmod-3.10 → genmod-3.10.2}/genmod/score_variants/score_function.py

@@ -29,8 +29,10 @@ class ModeLookup(Enum):
     STRING = 2
     # Score lookup in tree
     TREE = 3
+    # Score lookup is by determining if there is a value (flag set) or not
+    FLAG = 4
     # User provided score value
-    UNBOUNDED_USER_DEFINED = 4
+    UNBOUNDED_USER_DEFINED = 5
 
 
 class ScoreFunction(object):
@@ -211,7 +213,8 @@ class ScoreFunction(object):
         mode_value: bool = bool(self._value_dict)
         mode_str: bool = bool(self._string_dict)
         mode_tree: bool = bool(self._interval_tree)
-        if sum([mode_value, mode_str, mode_tree]) > 1:
+        mode_flag: bool = bool(self._reported_score)
+        if sum([mode_value, mode_str, mode_tree, mode_flag]) > 1:
             raise ValueError(
                 "Unable to accurately determine what mapping to use for determining score range"
             )
@@ -221,6 +224,8 @@ class ScoreFunction(object):
             return ModeLookup.STRING
         if mode_tree:
             return ModeLookup.TREE
+        if mode_flag:
+            return ModeLookup.FLAG
 
     @property
     def score_range(self) -> List[float]:
@@ -247,18 +252,20 @@ class ScoreFunction(object):
             scores: list = []
             for interval in self._interval_tree.all_intervals:
                 scores.append(interval.data)  # tree.interval -> score
+        elif self._scoring_mode == ModeLookup.FLAG:
+            scores: list = []
+            scores.append(float(self._reported_score))
         else:
             raise NotImplementedError("Unknown scoring mode", self._scoring_mode)
 
-        # Append set_reported and set_not_reported scores (as they're part of score value set)
         scores.append(float(self._not_reported_score))
-        scores.append(float(self._reported_score))
 
         if not isinstance(scores, list) and len(scores) > 0:
             raise KeyError("Found no score values", scores)
         for score_value in scores:
             if not isinstance(score_value, float):
                 raise TypeError("Invalid score type", score_value)
+
         return scores
 
     @property

{genmod-3.10 → genmod-3.10.2}/genmod/utils/variant_printer.py

@@ -12,12 +12,16 @@ Copyright (c) 2013 __MyCompanyName__. All rights reserved.
 from __future__ import print_function
 
 import logging
+import sys
 from codecs import open
-from multiprocessing import Process
+from multiprocessing import Process, log_to_stderr
 
 from genmod.utils import get_chromosome_priority, get_rank_score
 from genmod.vcf_tools import print_variant
 
+logger = log_to_stderr(logging.INFO)
+logging.basicConfig(stream=sys.stderr, force=True)
+
 
 class VariantPrinter(Process):
     """
@@ -45,7 +49,7 @@ class VariantPrinter(Process):
 
     def __init__(self, task_queue, head, mode="chromosome", outfile=None, silent=False):
         Process.__init__(self)
-        self.logger = logging.getLogger(__name__)
+        self.logger = logger
         self.task_queue = task_queue
         self.outfile = outfile
         self.header = head.header

{genmod-3.10 → genmod-3.10.2}/genmod/vcf_tools/sort_variants.py

@@ -39,7 +39,7 @@ def sort_variants(infile, mode="chromosome"):
     if mode == "chromosome":
         command.append("-k1,1V")  # Version sorting to deal with e.g. Un_* contigs
         command.append("-k3,3n")  # Sorting positions numerically
-        command.append("--stable")
+        command.append("-s")
 
     elif mode == "rank":
         command.append("-rn")

genmod-3.10.2/mkdocs.yml

@@ -0,0 +1,22 @@
+---
+site_name: genmod docs
+site_description: Project documentation with Markdown
+site_author: Måns Magnusson
+site_url: https://clinical-genomics.github.io/genmod
+repo_url: https://github.com/Clinical-Genomics/genmod
+
+nav:
+- Home: 'index.md'
+- Genetic Models: 'genetic-models.md'
+- Commands:
+    - Annotating Variants: 'commands/annotate-variants.md'
+    - Annotate Genetic Models: 'commands/annotate-models.md'
+    - Filter Variants: 'commands/filter-variants.md'
+    - Sort Variants: 'commands/sort-variants.md'
+    - Build References: 'commands/build-annotation.md'
+    - Score compounds: 'commands/score-compounds.md'
+    - Score variants: 'commands/score-variants.md'
+
+dev_addr: "0.0.0.0:4000"
+theme: mkdocs
+docs_dir: docs

{genmod-3.10 → genmod-3.10.2}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "genmod"
-version = "3.10"
+version = "3.10.2"
 description = "Annotate genetic inheritance models in Variant Call Format (VCF) files"
 authors = [{name="Mans Magnuson", email="mans.magnusson@scilifelab.se"}, {name="Tor Björgen", email="tor.bjorgen@scilifelab.se"}, {name="Felix Lenner", email="felix.lenner@scilifelab.se"}, {name="Anders Jemt", email="anders.jemt@scilifelab.se"}, {name="Daniel Nilsson", email="daniel.nilsson@ki.se"}]
 license = {text = "MIT License"}
@@ -33,7 +33,6 @@ dependencies = [
     "importlib_resources",
     "intervaltree >= 3.1.0",
     "extract_vcf",
-    "vcftoolbox",
     "six",
 ]
 
@@ -71,3 +70,8 @@ lint = [
     "isort>=5.11.5",
     "ruff>=0.8.0",
 ]
+docs = [
+	"mkdocs",
+	"mkdocs-material",
+	"markdown-include",
+]

{genmod-3.10 → genmod-3.10.2}/tests/score_variants/test_score_function.py

@@ -134,3 +134,24 @@ def test_score_mode_string_lookup():
     # THEN expect the proper min max values
     assert score_function.score_min == -100
     assert score_function.score_max == 2
+
+
+def test_score_mode_flag_lookup():
+    """Test ScoreFunctions min max bounds property."""
+    # GIVEN a score function with a flag reported/not_reported score
+    score_function = ScoreFunction(match_type="flag")
+    not_reported_score = -100
+    reported_score = 10
+    score_function.set_not_reported(not_reported_score)
+    score_function.set_reported(reported_score)
+
+    # WHEN accessing min max plugin score
+    # THEN expect the proper min max values
+    assert score_function.score_min == not_reported_score
+    assert score_function.score_max == reported_score
+
+    # WHEN asking for a missing value, THEN not reported is flagged and scored
+    assert score_function.get_score(None) == not_reported_score
+
+    # WHEN asking for a set value, THEN reported is flagged and scored
+    assert score_function.get_score("c'est une pipe") == reported_score