In silico candidate variant and gene identification using inbred mouse strains.

Area of Science:

• Genetics
• Bioinformatics
• Animal Models

Background:

• Mice are crucial animal models for studying genotype-phenotype relationships due to their genetic uniformity.
• Decades of phenotypic data and recent whole-genome sequencing data for inbred mouse strains are available.
• Identifying causal genetic variants for specific phenotypes remains a challenge.

Purpose of the Study:

• To develop and validate an in silico fine-mapping approach for identifying candidate variants and genes associated with observed phenotypes in inbred mice.
• To provide a user-friendly tool for researchers to analyze genotype and phenotype data.

Main Methods:

• Utilized genotypic data from 37 inbred mouse strains and user-provided phenotypic data.
• Developed a Bioconductor package, MouseFM, for efficient querying of 74 million variant sites.
• Filtered variants by molecular consequences and impact, generating candidate gene lists.
• Applied MouseFM to expression quantitative trait loci (eQTLs) datasets and specific phenotypes like albinism, interfrontal bone formation, and cardiac calcification.

Main Results:

• MouseFM successfully identified candidate variants and haplotypes for previously reported eQTLs in neutrophils and CD4+ T cells.
• For albinism, MouseFM pinpointed a single missense variant in the Tyr gene.
• Analysis of interfrontal bone formation yielded 12 candidate genes, three related to skull abnormalities.
• Dystrophic cardiac calcification analysis identified a single moderate impact variant in the Abcc6 gene.

Conclusions:

• MouseFM is an effective tool for in silico fine-mapping, enabling efficient identification of candidate variants and genes.
• The approach aids in dissecting genotype-phenotype relationships using readily available mouse data.
• MouseFM demonstrates utility across various complex traits and molecular data types.