GCPBayes pipeline: a tool for exploring pleiotropy at the gene level

Yazdan Asgari, Pierre Emmanuel Sugier, Taban Baghfalaki, Elise Lucotte, Mojgan Karimi, Mohammed Sedki, Amélie Ngo, Benoit Liquet, Thérèse Truong

Résultats de recherche: Contribution à un journalArticleRevue par des pairs

1 Citation (Scopus)

Résumé

Cross-phenotype association using gene-set analysis can help to detect pleiotropic genes and inform about common mechanisms between diseases. Although there are an increasing number of statistical methods for exploring pleiotropy, there is a lack of proper pipelines to apply gene-set analysis in this context and using genome-scale data in a reasonable running time. We designed a user-friendly pipeline to perform cross-phenotype gene-set analysis between two traits using GCPBayes, a method developed by our team. All analyses could be performed automatically by calling for different scripts in a simple way (using a Shiny app, Bash or R script). A Shiny application was also developed to create different plots to visualize outputs from GCPBayes. Finally, a comprehensive and step-by-step tutorial on how to use the pipeline is provided in our group's GitHub page. We illustrated the application on publicly available GWAS (genome-wide association studies) summary statistics data to identify breast cancer and ovarian cancer susceptibility genes. We have shown that the GCPBayes pipeline could extract pleiotropic genes previously mentioned in the literature, while it also provided new pleiotropic genes and regions that are worthwhile for further investigation. We have also provided some recommendations about parameter selection for decreasing computational time of GCPBayes on genome-scale data.

langue originaleAnglais
Numéro d'articlelqad065
journalNAR Genomics and Bioinformatics
Volume5
Numéro de publication3
Les DOIs
étatPublié - 1 sept. 2023
Modification externeOui

Contient cette citation