H3M2

H3M2 identifies runs of homozygosity (ROH) from whole-exome sequencing (WES) data to detect extended homozygous chromosomal segments relevant to population and medical genetics.

Key Features:

• Algorithmic Innovation: A heterogeneous hidden Markov model (HMM) that incorporates inter-marker distances to handle the sparse and non-uniform distribution of WES targets.
• Performance Evaluation: Benchmarking on synthetic chromosomes and data from the 1000 Genomes Project shows improved accuracy for detecting short, medium, and long ROHs compared with GERMLINE and PLINK.
• Software Composition: Implemented as scripts and code in bash, R, and Fortran.

Scientific Applications:

• Population Genetics: Investigating genetic structure and evolutionary history by characterizing ROH distributions from WES data.
• Medical Genetics: Identifying homozygous regions associated with genetic predispositions to rare and common disorders from WES data.

Methodology:

H3M2 adapts SNP-array analysis techniques within a heterogeneous HMM framework that integrates inter-marker distances to account for the non-uniform distribution of exome targets.