Git Repository Plink2SampleMetadata

PLINK to Sample Metadata Workflow

Overview

This Nextflow pipeline builds a sample metadata table from a PLINK binary dataset by performing:

• Sex call rate inference with PLINK
• Trio inference using KING
• Principal Component Analysis (PCA) using a KING reference panel
• Merging results into a single tab-delimited file (sample_metadata_from_plink.tsv)

Requirements

Refer to the template config files and adjust them to match your infrastructure.

Required software:

• Nextflow – workflow engine (nextflow version 25.10.2)
• Docker (Apptainer or Singularity) – to run containers

You might need to pull the following containers if working offline:

• docker://ghcr.io/jacquemontlab/plink2metadata:latest

Download required KING reference files

From the root directory of the repository, run the following command to save the KING reference data in ./resources/king_ref/ (plink and liftOver are provided by the Docker image):

docker run --rm -it \
  -v "$PWD":/project \
  -w /project \
  ghcr.io/jacquemontlab/plink2metadata:latest \
  bash INSTALL.sh

Inputs

Parameter	Description	Default
--plink_file	Path prefix of PLINK dataset (.bed/.bim/.fam)	Required
--king_ref	Path to KING reference directory	Required
--genome_version	Genome version for PCA analysis (GRCh37/GRCh38)	GRCh38

Usage

Testing

The pipeline can be tested using the test profile and the images hosted on github using the container platform of your choice.

container=docker # or apptainer or singularity

nextflow run main.nf -profile test,${container}

Example

plink_file=${PWD}/tests/plink

nextflow run main.nf \
    --plink_file "$plink_file" \
    --king_ref "$PWD"/resources/king_ref \
    --genome_version GRCh38

Outputs

The pipeline produces a merged metadata file: results/sample_metadata_from_plink.tsv

Below is a description of each column:

Column	Description
SampleID	Unique sample identifier from the PLINK dataset.
Call_Rate	Proportion of successfully called genotypes for the sample.
Sex	Inferred biological sex (male / female / unknown) based on PLINK sex check.
FatherID	Sample ID of the inferred father (if available).
MotherID	Sample ID of the inferred mother (if available).
FamilyID	Assigned family identifier for SampleID sharing same pair ofparents.
PC1–PC10	Principal component values from KING-based PCA analysis.
Ancestry	Inferred ancestry group based on PCA projection to the KING reference.

Example (first lines):

SampleID    Call_Rate   Sex    FatherID   MotherID   FamilyID   PC1    PC2   ...   PC10   Ancestry
ID1   0.97516     female                     -0.0101  0.0274 ... -0.0016  EUR
ID2   0.97283     male   ID3  ID7 Family2770 -0.0099  0.0272 ... -0.0017  EUR

Workflow Structure

Processes

• infer_sex_callrate Runs plink_sex_callrate.sh to estimate sex call rates per sample.

• infer_trio Uses infer_king_trios.py to infer trios from the genotype data.

• infer_pca Executes run_king_pca.sh with KING reference data for PCA analysis.

• merge_results Merges outputs into a single metadata file using merge_tsv.py.