ancQTL

ancQTL is an R script adapted from mixQTL for eQTL mapping. This version extends mixQTL by including local ancestry as an additional covariate, making it particularly useful for admixed populations where ancestry effects are of interest.

Features

• Based on the established mixQTL framework
• Adds support for local ancestry covariate
• Compatible with mixQTL input formats
• Suitable for local ancestry based (LAB) eQTL and sQTL analyses

Requirements

• R (>= 4.0)
• Data inputs compatible with mixQTL (see below)
• Local ancestry estimates per individual per gene

Installation

In R:

source("path/to/ancQTL/ancQTL.R")

Quick Start

The usage closely mirrors mixQTL, with one key addition: an input for local ancestry.

Input Data

• Genotype matrix (genotypes): Individuals × SNPs
• Expression vector (expression): Gene expression values for individuals
• Covariates matrix (covariates): Standard covariates (e.g., PCs, batch effects)
• Local ancestry vector (local_ancestry): Same dimension as genotypes, with local ancestry dosages

Example Usage

result <- ancqtl(
  geno1 = genotype1, 
  geno2 = genotype2,
  y1 = haplotype_specific_count1,
  y2 = haplotype_specific_count2,
  anc1 = local_ancestry1,
  anc2 = local_ancestry2,
  ytotal = total_read_count,
  lib_size = seq_lib_size,
  cov_offset = covariate_offset_matrix
)

The function returns a table of association results including local ancrstry based effect sizes and p-values.

Output

The output is a data frame with:

• SNP ID
• Effect size
• Standard error
• p-value
• Other relevant statistics

Differences from mixQTL

Feature	mixQTL	ancQTL
Standard covariates	✔	✔
Local ancestry	✖	✔