Lai, B. | Search bioRxiv

bioRxiv2021

Predicting Epigenomic Functions of Genetic Variants in the Context of Neurodevelopment via Deep Transfer Learning

Decoding the regulatory effects of non-coding variants is a key challenge in understanding the mechanisms of gene regulation as well as the genetics of common diseases. Recently, deep learning models have been introduced to predict genome-wide epigenomic profiles and effects of DNA variants, in various cellular contexts, but they were often trained in cell lines or bulk tissues that may not be related to phenotypes of interest. This is particularly a challenge for neuropsychiatric disorders, since the most relevant cell and tissue types are often missing in the training data of such models. To address this issue, we introduce a deep transfer learning framework termed MetaChrom that takes advantage of both a reference dataset - an extensive compendium of publicly available epigenomic data, and epigenomic profiles of cell types related to specific phenotypes of interest. We trained and evaluated our model on a comprehensive set of epigenomic profiles from fetal and adult brain, and cellular models representing early neurodevelopment. MetaChrom predicts these epigenomic features with much higher accuracy than previous methods, and than models without the use of reference epigenomic data for transfer learning. Using experimentally determined regulatory variants from iPS cell-derived neurons, we show that MetaChrom predicts functional variants more accurately than existing non-coding variant scoring tools. By combining genome-wide association study (GWAS) data with MetaChrom predictions, we prioritized 31 SNPs for Schizophrenia (SCZ). These candidate SNPs suggest potential risk genes of SCZ and the biological contexts where they act. In summary, MetaChrom is a general transfer learning framework that can be applied to the study of regulatory functions of DNA sequences and variants in any disease-related cell or tissue types. The software tool is available at https://github.com/bl-2633/MetaChrom and a prediction web server is accessible at https://metachrom.ttic.edu/.

bioinformatics↗

bioRxiv2020

Reversing Transcriptome-Wide Association Studies to improve expression Quantitative Trait Loci associations

1Transcriptome-Wide Association Studies discover SNP effects mediated by gene expression through a two-stage process: a typically small reference panel is used to infer SNP-expression effects, and then these are applied to discover associations between imputed expression and phenotypes. We investigate whether the accuracy of SNP-expression and expression-phenotype associations can be increased by performing inference on both the reference panel and independent GWAS cohorts simultaneously. We develop EMBER (Estimation of Mediated Binary Effects in Regression) to re-estimate these effects using a liability threshold model with an adjustment to variance components accounting for imputed expression from GWAS data. In simulated data with only gene-mediated effects, EMBER more than doubles the performance of SNP-expression linear regression, increasing mean r2 from 0.3 to 0.65 with a gene-mediated variance explained of 0.01. EMBER also improves estimation accuracy when the fraction of cis-SNP variance mediated by genes is as low as 30%. We apply EMBER to genotype and gene expression data in schizophrenia by combining 512 samples from the CommonMind Consortium and 56,081 samples from the Psychiatric Genomic Consortium. We evaluate performance of EMBER in 36 genes suggested by TWAS by concordance of inferred effects with effects reported independently for frontal cortex expression. Applying the EMBER framework to a baseline linear regression model increases performance in 26 out of 36 genes (sign test p-value .0020) with an increase in mean r2 from 0.200 to 0.235.

genomics↗