Skip to content

Latest commit

 

History

History
70 lines (41 loc) · 3.39 KB

README.md

File metadata and controls

70 lines (41 loc) · 3.39 KB

GPScore

The Gene Priority Score (GPScore) is a combinatorial likelihood scoring formalism based on measures derived from 11 gene prioritization strategies and the physical distance to the transcription start site (TSS). Using GPScore, you can prioritize the most probable target genes underlying the GWAS-associated variants.

Accompanying paper:

Vishal Sarsani, Sarah M Brotman, Xianyong N Yin, Lilian Fernandes Silva, Markku Laakso, Cassandra N Spracklen. A multi-ancestry genome-wide meta-analysis, fine-mapping, and target gene prioritization to characterize the genetic architecture of adiponectin. medRxiv 2023.05.02.23289402; doi: https://doi.org/10.1101/2023.05.02.23289402

Overview

Genome-Wide Association Studies(GWAS) for complex traits typically result in many significant loci across the genome. Translating these GWAS findings into treatments may require an understanding of target genes and their biological mechanisms.

The genetic variants linked to a certain trait may not always affect the closest gene and can even impact protein levels located far away. Various methods exist to identify the genes affected by these variants, but they have limitations, such as not being customizable to relevant disease/tissue data and producing conflicting results. GPScore combines multiple gene prioritization strategies and the physical distance to transcription start sites, allowing for customization and unbiased scoring. It can be applied to complex traits with limited training data and does not have individual-level data.

Gpscore

Data and software

Outputs from following software/data should be available in text or csv format to calculate the GPScore:

The following R (version 4.2.0) packages are required:

Example dataset

A dataset is taken from Asian Genetic Epidemiology Network (AGEN : https://blog.nus.edu.sg/agen/). GWAS meta-analysis (imputed to HapMap2) for adiponectin levels in up to 7,825 East Asians.

zcat data/AGEN_adiponectin_hapmap.txt.gz|head -5

Usage and Tutorial

The main notebook to calculate GPScore is GPscore.ipynb.

Several other notebooks exists to help process the data from other tools

Authors

License

This software is distributed under the GPLv3 license.