senresearch / FlxQTL.jl

A comprehensive QTL analysis tool by multivariate linear mixed model.

FlxQTL

Flexible Multivariate Linear Mixed Model based QTL Analysis for Structured Multiple Traits

FlxQTL.jl is a package for a multivariate linear mixed model based QTL analysis tool that supports incorporating information from trait covariates such as time or different environments. The package supports computation of one-dimensional and two-dimensional multivariate genome scans, visualization of genome scans, support for LOCO (leave-one-chromosome-out), computation of kinship matrices, and support for distributed computing.

The package is written in Julia and includes extensive documentation. If you are new to Julia you may want to learn more by looking at Julia documentation. Example data sets are located in the data directory. For details about the method, you may want to read our paper available as a preprint.

Paper

Flexible multivariate linear mixed models for structured multiple traits Hyeonju Kim, Gregory Farage, John T. Lovell, John K. Mckay, Thomas E. Juenger, Śaunak Sen doi: https://doi.org/10.1101/2020.03.27.012690

Installation

The package can installed in following ways. In a Julia REPL, press ] to enter a package mode,

julia> ]
pkg> add FlxQTL

Or, equivalently,

julia> using Pkg; Pkg.add("FlxQTL")

For installing from the source,

pkg> add https://github.com/senresearch/FlxQTL.jl

or,

julia> Pkg.add(url="https://github.com/senresearch/FlxQTL.jl")

To remove the package from the Julia REPL,

julia> ]
pkg> rm FlxQTL

Equivalently,

julia> using Pkg; Pkg.rm("FlxQTL")

Choice of BLAS vendors

The package can be run with OpenBLAS (built-in Julia dense linear algebra routines) or MKL (Intel's Math Kernel Library). MKL.jl works best on Intel hardware, but it can slightly improve performance without Intel hardware. For installation and details, see: MKL.jl.