pmbb-geno-pheno-toolkit

This repository has an overview of our sub-repositories for each of the modules we have developed within the PMBB-Informatics working group. Contents include:

• Links to repos
• Example config files
• Set-up instructions

Our Pipelines:

Currently Published + Cluster and AOU-Tested:

• GWAS with PLINK 2.0
• GWAMA for GWAS Meta-Analysis
• Meta-Analysis for ExWAS Results

Currently Published + Cluster-Tested:

• SAIGE FAMILY
  • ExWAS
  • GWAS
  • Gene Burden PheWAS
  • Single Variant PheWAS
• PRS-CSx for Multi-Ancestry PRS Development

Coming Soon:

• Compute PGS with Plink --score
• PLINK Clump
• MetaXcan Pipelines

Under Construction:

• LDSC Pipelines
  • Genetic Correlation
  • Heritability Estimation
  • Partitioned Heritability Estimation
• OMOP-based simple phenotyping

Getting Started:

Things to install/download:

• Nextflow (version 23 is ideal)
• Docker OR Singularity to make use of our containers
• Other pipeline-specific things you may need to build/download, such as:
  • LD Panels for PRS-CSx
  • Loki for biofilter annotations

How to Start Working with a Pipeline:

1. Clone the sub-module for the pipeline you need
2. Edit the pipeline-specific .config file to include your parameters and input data
3. Set up a nextflow.config file with a profile for your compute system. We have an example for this as well HERE.
4. Pro Tip: at the top of your nextflow.config, add a line: includeConfig {pipeline}.config with the relative path to your new .config file. The alternative is using the -c / -config flag when calling the pipeline.
5. In order to run one of the workflows, you will use a command like this: nextflow run /path/to/pmbb-nf-toolkit-{pipeline}/{pipeline}.nf. More details can be found in module-specific READMEs

Resources:

• PMBB Website
• All of Us Home
• nf-core Getting Started guide
• Nextflow-provided training
• Nextflow Hello World and Intermediate Crash Course (Google CoLab)
• Nextflow Hello Penguins: Real Data Exaple