Premium
FastLORS: Joint modelling for expression quantitative trait loci mapping in R
Author(s) -
Rhyne Jacob,
Jeng X. Jessie,
Chi Eric C.,
Tzeng JungYing
Publication year - 2020
Publication title -
stat
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.61
H-Index - 18
ISSN - 2049-1573
DOI - 10.1002/sta4.265
Subject(s) - expression quantitative trait loci , computer science , coordinate descent , multivariate statistics , quantitative trait locus , software , expression (computer science) , block (permutation group theory) , computational statistics , data mining , algorithm , gradient descent , artificial intelligence , mathematics , machine learning , biology , combinatorics , genetics , gene , genotype , single nucleotide polymorphism , artificial neural network , programming language
FastLORS is a software package that implements a new algorithm to solve sparse multivariate regression for expression quantitative trait loci (eQTLs) mapping. FastLORS solves the same optimization problem as LORS, an existing popular algorithm. The optimization problem is solved through inexact block coordinate descent with updates by proximal gradient steps, which reduces the computational cost compared with LORS. We apply LORS and FastLORS to a real dataset for eQTL mapping and demonstrate that FastLORS delivers comparable results with LORS in much less computing time.