Merge pull request #914 from uclahs-cds/czhu-fix-mito

Support Different Codon Tables

Author: Chenghao (Trevor) Zhu

.gitignore

@@ -18,4 +18,6 @@ build
 
 test/files
 
-venv
+venv
+
+notebooks

CHANGELOG.md

@@ -10,6 +10,34 @@ This project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.htm
 
 ## [Unreleased]
 
+## [1.5.0] - 2025-06-15
+
+- Added `--codon-table` and `--chr-codon-table`. The former sets the codon table to use, and the latter overrides it for specific chromosomes.
+
+- Added `--star-codons` and `--chr-star-codons` to specify start codons to use.
+
+- Added the support for codon table for `callVariant`.
+
+- Added the support for codon table for `callNovelORF` and `callAltTranslation`.
+
+- Added codon table to `downsampleReference`.
+
+- Fixed graph algorithms to use specified codon table and start codon
+
+- Fixed `GenomicAnnotationOnDisk` that when using on-the-fly indices, the `is_protein_coding` attributes of transcripts are not updated correctly.
+
+- Added `force_init_met` to `PeptideVariantGraph.call_variant_peptide` to control whether the initial amino acid should be forced to Methionine.
+
+- Updated the reference data loading function to directly return a `ReferenceData` object.
+
+- Updated `bruteForce` to specify codon table and start codons.
+
+- Updated `fuzzTest` to pass codon tabel and start codons to `callVariant` and `bruteForce`
+
+- Fixed `callVariant` that during TVG alignment, node merged from multiple frameshifts which together go back to the origianl frame, was not recognized as `was_brige` in a bubble
+
+- Fixed `fuzzTest` that cds start should be at least 3 nucleotide away from the cds end.
+
 ## [1.4.6] - 2025-05-21
 
 - Fixed biopython version #908

docs/call-alt-translation.md

@@ -11,6 +11,10 @@
 
 {% include 'partials/_caution_on_reference_version.md' %}
 
+{% include 'partials/_args_reference.md' %}
+
+{% include 'partials/_args_codon_table.md' %}
+
 ## Usage
 
 ```

docs/call-novel-orf.md

@@ -11,6 +11,10 @@
 
 {% include 'partials/_caution_on_reference_version.md' %}
 
+{% include 'partials/_args_reference.md' %}
+
+{% include 'partials/_args_codon_table.md' %}
+
 ## Usage
 
 ```

docs/call-variant.md

@@ -9,6 +9,10 @@
       show_root_heading: false
       show_source: false
 
+{% include 'partials/_args_reference.md' %}
+
+{% include 'partials/_args_codon_table.md' %}
+
 ## Usage
 
 ```

docs/files/fuzz_test_history.tsv

@@ -41,3 +41,6 @@ v1.4.5	f46742d	2025-02-24	comprehensive	5902	0	0	0:00:00.375135	1.01575471228857
 v1.4.6-rc1	78e971d	2025-03-04	snv	2710	0	0	0:00:00.163103	0.38221870483830245	0:00:56.301104	116.21712282069569
 v1.4.6-rc1	78e971d	2025-03-04	indel	2850	0	0	0:00:00.191917	0.3938244401664319	0:00:41.244215	95.36212234507254
 v1.4.6-rc1	78e971d	2025-03-04	comprehensive	5310	0	0	0:00:00.395482	1.0924741762245143	0:00:40.031276	179.44714992445952
+v1.4.6-rc4	8f8e871	2025-06-14	snv	3899	0	0	0:00:00.161568	0.37222759958119433	0:00:55.336499	116.08301043924084
+v1.4.6-rc4	8f8e871	2025-06-14	indel	3811	0	0	0:00:00.253212	1.900065095497631	0:00:39.172671	89.15186913116658
+v1.4.6-rc4	8f8e871	2025-06-14	comprehensive	7291	0	0	0:00:00.388579	0.9478303137705371	0:00:42.034437	171.5405703540714

docs/partials/_args_codon_table.md

@@ -0,0 +1,56 @@
+## Codon Table & Start Codon
+
+### Codon Table
+
+The NCBI standard codon table is used by default, which is used for the majority of nulcear gene translation in eukaryote cells. See [here](https://www.ncbi.nlm.nih.gov/Taxonomy/Utils/wprintgc.cgi) for a complete list of NCBI codon tables.
+
+The default codon table can be override using `--codon-table`. For example:
+
+```shell
+--codon-table ’Ciliate Nuclear'
+```
+
+The `--chr-codon-table` can be used to specify the codon table used for a specific chomosome. The example below uses the 'Vertebrate Mitochondrial' (SGC1) codon table for genes from the mitochondria chomosome, and the standard codon table otherwise.
+
+```shell
+--codon-table Standard \
+--chr-codon-table 'chrM:SGC1'
+```
+
+### Start Codons
+
+Stard codons usually do not need to be specified. The standard start codon `ATG` is used by default, and it is translated as Methionine as start codon and in elongation. However, in some cases, for example, mitochondria, `ATA` and `ATT` may also be used as start codon. While `ATT` is translatted into Isoleucine during elongation, Methionine is still used as start codon.
+
+Similar to codon table, the default codon table can be override using `--start-codons`.
+
+```shell
+--start-codons ATG
+```
+
+The `--chr-start-codon` can also be used to assign start codons to a specific chomosome. The example below assigns `ATG`, `ATA`, and `ATT` to the mitochondrial chromosome.
+
+```shell
+--chr-start-codons 'chrM:ATG,ATA,ATT'
+```
+
+### Default
+
+The chromosome names must be specified correctly, same as what used in the genome fasta and annotation GTF file. By default, moPepGen infers the reference source of the annotation (*i.e.*, GENCODE or EMSEMBL), and uses the 'SGC1' codon table for mitochondirla chromosome. So the default is equivalent to:
+
+```shell
+--reference-source GENCODE \
+--codon-table Standard \
+--chr-codon-table 'chrM:SGC1' \
+--start-codons 'ATG' \
+--chr-start-codongs 'chrM:ATG,ATA,ATT
+```
+
+or
+
+```shell
+--reference-source ENSEMBL \
+--codon-table Standard \
+--chr-codon-table 'MT:SGC1' \
+--start-codons 'ATG' \
+--chr-start-codongs 'MT:ATG,ATA,ATT
+```

docs/partials/_args_reference.md

@@ -0,0 +1,8 @@
+## Reference
+
+Reference data, incluiding reference genome, genome annotation, and protein coding translation are required. There are two ways of specifying reference data:
+
+1. Using the index dir created by the [`generateIndex`](/generate-index) command.
+2. Specifying each reference files needed.
+
+1 is highly recommended as it is faster and helps you ensure that the same reference data are used across the project.

moPepGen/__init__.py

@@ -8,7 +8,7 @@
 from . import constant
 
 
-__version__ = '1.4.6'
+__version__ = '1.5.0'
 
 ## Error messages
 ERROR_INDEX_IN_INTRON = 'The genomic index seems to be in an intron'

moPepGen/aa/AminoAcidSeqRecord.py

@@ -335,10 +335,6 @@ def find_all_enzymatic_cleave_sites_with_ranges(self, rule:str, exception:str=No
         return list(self.iter_enzymatic_cleave_sites_with_range(rule=rule,
             exception=exception))
 
-    def find_all_start_sites(self) -> List[int]:
-        """ Find all start positions """
-        return [x.start() for x in re.finditer('M', str(self.seq))]
-
     def find_all_cleave_and_stop_sites(self, rule:str, exception:str=None,
             exception_sites:List[int]=None) -> List[int]:
         """ Find all enzymatic lceave sites and stop sites """