Skip to contents

Generate mapping data and peaks for QTL analysis

mapQTL.Rd

This function performs QTL mapping across multiple tissues, generating genome-wide LOD scores and identifying significant peaks. It supports parallel processing and flexible input formats for expression data and covariates.

Usage

mapQTL(
  genoprobs,
  samp_meta,
  expr_mats,
  covar_factors,
  thrA = 5,
  thrX = 5,
  gridFile = gridfile,
  localRange = 2e+06,
  outdir = NULL,
  peaks_out = "peaks.rds",
  map_out = "map.rds",
  annots = NULL,
  total_cores = NULL,
  save = "sr",
  rz = FALSE,
  phys = TRUE,
  ...
)

Arguments

genoprobs

Genotype probabilities object (qtl2-formatted), or path to .rds file containing one.

samp_meta

Sample metadata. Either a string pointing to the file, or the object itself.

expr_mats

List of normalized count matrices (objects), or character paths to the file. One matrix per tissue. The order must match the tissue order in genoprobs. NOTE If using rz = TRUE then phenotypes should be the columns in the expression matrices.

covar_factors

Additive covariate factors. These need to be columns in the factor metadata.

thrA

Minimum reported LOD threshold for autosomes. Default is 5.

thrX

Minimum reported LOD threshold for X chromosome. Default is 5.

gridFile

Genome Grid. Path to location or object. Defaults to 75k grid loaded with package.

localRange

Definition of "local" in bp. Default is 2e6.

outdir

Directory to save output files. Default is NULL.

peaks_out

String indicating the name for output peaks file. This file will be saved in .rds format and will be used as an input for downstream analysis. Should end in .rds. Default is "gxe_peaks.rds"

map_out

String indicating the name of the output file containing QTL mapping results. This file will be saved in .rds format and will be used in downstream analyses. Should end in .rds. Default is "map.rds"

annots

Annotations file. Contains mapping information for phenotypes. Dataframe, or tsv. Columns must include "id", "symbol", "chr", "start", "end".

total_cores

Number of available cores to use for parallelization. Default is NULL.

save

Indicates object return/save behavior. One of c("sr", "so", "ro"); save & return, save only, return only. Default is "sr".

rz

Logical. Set to TRUE if expression data is already rankZ-transformed. Default is FALSE.

phys

Logical. if TRUE, use the physical map for peak calling; otherwise use the genomic map. Default is TRUE.

Value

A list containing:

  • maps_listA list of dataframes and lists to that can be used for future analyses and in other functions

    • qtlprobsGenome probabilities in qtl2format

    • covar_listlist of covariate matrices for each tissue

    • expr_listOriginal normalized expression data for each tissue

    • exprZ_listRank Z-transformed expression data for each tissue

    • kinship_locoKinship matrix calculated using the "loco" option in qtl2::calc_kinship

    • gmapGenomic map of markers

    • map_dat2Combined genomic and physical map of markers

    • pmapPhysical map of markers

    • tissue_sampMetadata broken down for each tissue

  • peaks_listA list of peaks list for each tissue.