Research Alert


FOR IMMEDIATE RELEASE: Nature Communications: August 23, 2019

Integrative transcriptome imputation reveals tissue-specific and shared biological mechanisms mediating susceptibility to complex traits

Corresponding Author:  Panos Roussos, MD, PhD, Associate Professor of Genetics and Genomic Sciences and Psychiatry, Icahn School of Medicine at Mount Sinai

Bottom Line: We can leverage genetic studies and large-scale omics data (novel, comprehensive approaches for analysis of complete genetic or molecular profiles of humans and other organisms) in human tissues to model tissue-specific gene expression dysregulation in complex traits and subsequently identify compounds that would reverse these genetically driven perturbations.

Results: By developing an improved method for predicting the effects of genetic variation on gene expression we were able to better study the interplay between complex diseases and identify existing drugs that could be explored for new medical indications.

Why the Research Is Interesting: As large -omics data become increasingly available, it is important to develop methods that will integrate them with both individual and population-level genetic data for the advent of personalized and precision medicine applications, respectively.

Who: The study was led by researchers at The Roussos Lab at Mount Sinai and carried out by a cross-disciplinary, multi-site team that specializes in the generation of large-scale molecular data in human tissues and their subsequent integration with genetic risk factors to identify novel druggable targets.

When: The study started 3 years ago.

What: Joint analysis of 14 large-scale human tissue transcriptome datasets, 58 genome wide association studies for human traits and tissue-specific epigenetic annotations. Our gene expression prediction method modeled the genetically driven gene expression changes of these 58 traits across 8 tissue types and computational drug repurposing analysis revealed candidate compounds that could reverse them.

How: By developing an improved method for predicting the effects of genetic variation on gene expression (transcriptomic imputation) which incorporates epigenome data, we were able to integrate genetic, epigenetic and gene expression information to identify genetically driven gene expression changes in several human diseases. Then, by using reference panels of the effects of existing drugs on gene expression, we were able to identify compounds that could be therapeutic by countering these changes.

Study Conclusions: Methods that can model the effect of genetic variation on gene expression can increase the translational opportunities of existing genetic studies and pave the way for personalized and precision medicine.

Said Mount Sinai's Dr. Panos Roussos of the research: "This study is part of our long-term efforts to identify translational opportunities of genetic, epigenetic and genomic findings. As a proof of concept, we demonstrate that the integration of these -omics datasets can identify candidate compounds that can be explored for new medical indications - an approach that holds promise for personalized and precision medicine."