How to adjust PGS in the context of genetic ancestry

pgscatalog-ancestry-ajust is a CLI application that adjusts calculated PGS using genetic ancestry similarity data.

The process requires some complex inputs. If you’d like to automatically adjust PGS in the context of genetic ancestry automatically, it’s probably simplest to use the PGS Catalog Calculator, which uses this application internally.

Installation

$ pip install pgscatalog-calc

Usage

The following inputs are needed:

• PGS calculated on a target and reference population in long (melted) format (--agg_scores)

  • PGS must be calculated using the same set of variants (the intersection)

  • Input file structure is the same as the output from pgscatalog-aggregate

• PCA derived from a reference population using fraposa-pgsc (--ref_pcs)

• PCA for target population projected onto reference PCA using fraposa-pgsc (--target_pcs)

• A relatedness cutoff file for the reference population (--reference_related)

• A sample information file for the reference population containing genetic ancestry group labels (--psam, SuperPop column by default)

• Text labels of the reference and target samples (--reference and --dataset), which must match the aggregated scoring file contents

$ pgscatalog-ancestry-adjust --agg_scores aggregated_scores.txt.gz \
    --psam reference.psam \
    --ref_pcs ref.pcs \
    --target_pcs target.pcs \
    --reference_related ref.king.cutoff.id \
    --dataset hgdp \
    --reference reference \
    --outdir ancestry_results

Help

$ pgscatalog-ancestry-adjust --help