bioh2m 1.0.3

For precision modeling of human variants in the mouse genome.

H2M (Human-To-Mouse)

Author: Kexin Dong
Date: May 2, 2024

---

H2M is a Python package for the precision modeling of human variants in mice and vice versa.

H2M's main functions are:

1. Reading and formatting mutation data from different pulic sources.

2. Querying orthologous genes between mouse and human.

3. Generating murine equivalents for human genetic variant input or vice versa.

See more in the the GitHub repository.

Complete documentation at H2M Readthedocs.

Installation

1. Via PyPI

pip install bioh2m

2. Download the .whl file from the GitHub repository

pip install bioh2m-1.0.0-py3-non-any.whl  

ATTENTION: H2M has pysam as a dependency. This is for a function that can read .vcf files. If you are experiencing installation problems due to pysam, you can download and install the wheel file in the GitHub repository without this function and the pysam dependency, which has been tested to solve most installation issues. The function rounded off in mini-h2m is also given in the repo.

H2M has been tested in Python 3.9-3.12.

Download dependency files

Reference genome and genome annotation files for human and mouse are needed for this package. To download, visit the Dropbox Folder. Some commonly used public data of human mutations are also available here.

Quick Start

Load reference genome

import bioh2m as h2m
import pandas as pd

path_h_ref, path_m_ref = '.../GCF_000001405.25_GRCh37.p13_genomic.fna.gz', '.../GCF_000001635.27_GRCm39_genomic.fna.gz'
# remember to replace the paths with yours; for human, GRCh38 reference genome assembly is also provided  
records_h, index_list_h = h2m.genome_loader(path_h_ref)
records_m, index_list_m  = h2m.genome_loader(path_m_ref)

path_h_anno, path_m_anno = '.../gencode_v19_GRCh37.db', '.../gencode_vm33_GRCm39.db'
# remember to replace the paths with yours
db_h, db_m = h2m.anno_loader(path_h_anno), h2m.anno_loader(path_m_anno)

Load human mutation data

filepath = '/Users/kexindong/Documents/GitHub/Database/PublicDatabase/ClinVar/GRCh37_clinvar_20240206.vcf.gz'
variation_ids = [32798013, 375926, 325626, 140953, 233866, 1796995, 17578, 573320]
df = h2m.clinvar_reader(filepath, variation_ids)
df = h2m.clinvar_to_maf(df)
df = df[['gene_name_h',	'start_h','end_h','ref_seq_h','alt_seq_h','type_h','format','ID']]
df = df.rename(columns={'ID':'index'})

Get canonical transcript IDs for the human genes

df, df_fail = h2m.get_tx_batch(df, species='h', ver = 37)

Query the gene orthologs in mouse

df_queried, df_fail = h2m.query_batch(df, direction='h2m')

Get canonical transcript IDs for the murine genes

df_queried, df_fail = h2m.get_tx_batch(df_queried, species='m')

Compute the muerine variant equivalents

df_result, df_fail = h2m.model_batch(df_queried, records_h, index_list_h, records_m, index_list_m, db_h, db_m, 37)

Data visualization

df_vs = h2m.model(records_h, index_list_h, records_m, index_list_m, db_h, db_m, 
                  tx_id_h='ENST00000349496.5',tx_id_m='ENSMUST00000007130.15',
                  start=41266098, end=41266098, ref_seq='A', alt_seq='G', ty_h = 'SNP',
                  ver = 37, show_sequence=True)
h2m.visualization(df_vs, flank_size=2, print_size=6)

License

h2m is distributed under the terms of the MIT license.