Welcome to INFIMA!

INFIMA is an R package for the Integrative Fine-Mapping with Model Organism Multi-Omics Data. INFIMA utilizes the diversity outbred (DO) mice population as the model organsim. The major usage of the INFIMA package is to fine-map the eQTL markers in DO studies (DO-eQTL). INFIMA implements an empirical Bayes model which quantifies how well each non-coding SNP explains the observed DO allelic patern through consistency of founder mice RNA-seq data, founder mice ATAC-seq data (including the existence and consistency of a footprint) with the observed allelic pattern.

What do we obtain from INFIMA?

Given a library of mouse SNPs, the corresponding local ATAC-seq signals, DO-eQTL data as well as the gene expression in founder mice, INFIMA provides the following functionalities:
(1) Fine-mapping DO-eQTLs and estimate SNP-level posterior probabilities.
(2) Linking SNPs or the local ATAC-seq peaks to effector genes.

How do we utilize INFIMA predictions for human GWAS?

INFIMA results from DO studies can be mapped to human orthologs using peak-based lift-over strategies, which provides putative effector genes of human GWAS SNPs. See the "GWAS Effector Genes" page for the results validated by promoter capture Hi-C data.

About

Email us: Chenyang Dong, Sunduz Keles
Copyright (c) Chenyang Dong and Sunduz Keles
INFIMA is a collaboration between the Keles and Attie Labs.

INFIMA model overview

Effector genes of islet human GWAS SNPs validated by pcHi-C

INFIMA Predictions

DO mouse QTL markers were fine-mapped to local-ATAC-MVs by using INFIMA.

Click any row of the following table to see the relevant multi-omics data in the "Data Visualization" page.

Data Visualization

Selected INFIMA prediction
Genotype of the local-ATAC-MV
Posterior probability of the local-ATAC-MV