HTRX: an R package for learning non-contiguous haplotypes associated with a phenotype

Area of Science:

• Genetics
• Bioinformatics
• Statistical genomics

Background:

• Genome-wide association studies (GWAS) identify numerous SNPs linked to complex traits, but pinpointing causal variants within linkage disequilibrium blocks remains difficult.
• Understanding genetic architecture requires assessing main effects, interactions, and tagging effects of genetic variants.
• Haplotype-based approaches offer a powerful framework for dissecting complex genetic signals.

Purpose of the Study:

• To introduce Haplotype Trend Regression with eXtra flexibility (HTRX), an R package designed to learn sets of interacting features explaining phenotypic variance.
• To enable the quantification of total variance explained by main effects, interactions, and tagging effects using haplotype associations.
• To provide a computationally efficient tool for analyzing large chromosomal regions in genetic studies.

Main Methods:

• HTRX utilizes haplotype-based associations to identify sets of interacting genetic features (haplotypes) associated with a phenotype.
• The package incorporates strategies like 'Cumulative HTRX' to constrain feature set search space and manage computational complexity.
• Computational time scales linearly with the number of SNPs, allowing application to extensive genomic regions.

Main Results:

• HTRX identifies haplotypes, including those composed of non-contiguous SNPs, that are associated with phenotypic variation.
• The package facilitates the analysis of genetic regions with existing GWAS hits, either before or after fine-mapping.
• The linear scaling of computation time with SNP number makes HTRX suitable for large-scale genomic analyses.

Conclusions:

• HTRX provides a flexible and computationally efficient R package for identifying interacting genetic features and quantifying their contribution to phenotypic variance.
• The package is valuable for dissecting complex genetic architectures by considering haplotype effects beyond single SNPs.
• HTRX offers a promising tool for genetic research, particularly in analyzing large genomic regions and complex traits.