Speeding up eQTL scans in the BXD population using GPUs

Abstract

The BXD family of mouse strains are an important reference population for systems biology and genetics that have been fully sequenced and deeply phenotyped. To facilitate interactive use of genotype-phenotype relations using many massive omics data sets for this and other segregating populations, we have developed new algorithms and code that enable near-real-time whole genome QTL scans for up to one million traits. By using easily parallelizable operations including matrix multiplication, vectorized operations, and element-wise operations, our method is more than 700 times faster than a R/qtl linear model genome scan using 16 threads. We used parallelization of different CPU threads as well as GPUs. We found that the speed advantage of GPUs is dependent on problem size and shape (the number of cases, number of genotypes, and number of traits). Our approach is ideal for interactive web services, such as GeneNetwork.org that need to display results in real-time. Our implementation is available as the Julia language package LiteQTL at https://github.com/senresearch/LiteQTL.jl.

Genetic Modulation of Protein Expression in Rat Brain

“…The endophenotypes encompass a wide spectrum of molecules—mRNAs, proteins, and metabolites—that serve as intermediates in complex cascades that link DNA variants to higher order traits. The accuracy and throughput with which molecular endophenotypes can now be quantified in larger populations and families holds significant promise in bridging the gap between genomic loci and traits 7,8 .…”

“…The endophenotypes encompass a wide spectrum of molecules-mRNAs, proteins, and metabolites-that serve as intermediates in complex cascades that link DNA variants to higher order traits. The accuracy and throughput with which molecular endophenotypes can now be quantified in larger populations and families holds significant promise in bridging the gap between genomic loci and traits 7,8 . Almost all work on the genetics of expression and its linkage to higher order traits has been involved in mRNA expression first using arrays and now using RNA sequencing [9][10][11] .…”

BulkLMM: Real-time genome scans for multiple quantitative traits using linear mixed models

“…The main reason for the speedup is that there are efficient algorithms for matrix multiplication. The ten-sorQTL (Taylor-Weiner et al ., 2019) and LiteQTL (Trotter et al ., 2021) packages used GPUs to speed up the computations further. They use the fact that matrix multiplication uses similar operations with different data, which is an ideal candidate for GPU computation.…”