Abstract:The genome-wide association studies (GWAS) have identified hundreds of susceptibility loci associated with autoimmune diseases. However, over 90% of risk variants are located in the noncoding regions, leading to great challenges in deciphering the underlying causal functional variants/genes and biological mechanisms. Previous studies focused on developing new scoring method to prioritize functional/disease-relevant variants. However, they principally incorporated annotation data across all cells/tissues while omitted the cell-specific or context-specific regulation. Moreover, limited analyses were performed to dissect the detailed molecular regulatory circuits linking functional GWAS variants to disease etiology. Here we devised a new analysis frame that incorporate hundreds of immune cell-specific multi-omics data to prioritize functional noncoding susceptibility SNPs with gene targets and further dissect their downstream molecular mechanisms and clinical applications for 19 autoimmune diseases. Most prioritized SNPs have genetic associations with transcription factors (TFs) binding, histone modification or chromatin accessibility, indicating their allelic regulatory roles on target genes. Their target genes were significantly enriched in immunologically related pathways and other immunologically related functions. We also detected long-range regulation on 90.7% of target genes including 132 ones exclusively regulated by distal SNPs (eg, CD28, IL2RA ), which involves several potential key TFs (eg, CTCF), suggesting the important roles of long-range chromatin interaction in autoimmune diseases. Moreover, we identified hundreds of known or predicted druggable genes, and predicted some new potential drug targets for several autoimmune diseases, including two genes ( NFKB1, SH2B3 ) with known drug indications on other diseases, highlighting their potential drug repurposing opportunities. In summary, our analyses may provide unique resource for future functional follow-up and drug application on autoimmune diseases, which are freely available at <http://fngwas.online/>. Author Summary Autoimmune diseases are groups of complex immune system disorders with high prevalence rates and high heritabilities. Previous studies have unraveled thousands of SNPs associated with different autoimmune diseases. However, it remains largely unknown on the molecular mechanisms underlying these genetic associations. Striking, over 90% of risk SNPs are located in the noncoding region. By leveraging multiple immune cell-specific multi-omics data across genomic, epigenetic, transcriptomic and 3D chromatin interaction information, we systematically analyzed the functional variants/genes and biological mechanisms underlying genetic association on 19 autoimmune diseases. We found that most functional SNPs may affect target gene expression through altering transcription factors (TFs) binding, histone modification or chromatin accessibility. Most target genes had known immunological functions. We detected prevailing long-range chromatin interaction linking distal functional SNPs to target genes. We also identified many known drug targets and predicted some new drug target genes for several autoimmune diseases, suggesting their potential clinical applications. All analysis results and tools are available online, which may provide unique resource for future functional follow-up and drug application. Our study may help reduce the gap between traditional genetic findings and biological mechanistically exploration of disease etiologies as well as clinical drug development.

sscNOVA: a semi-supervised convolutional neural network for predicting functional regulatory variants in autoimmune diseases

Diagnostic Classification and Prognostic Prediction Using Common Genetic Variants in Autism Spectrum Disorder: Genotype-Based Deep Learning

Assessing concordance among human, in silico predictions and functional assays on genetic variant classification.

DeepWAS: Multivariate Genotype-Phenotype Associations by Directly Integrating Regulatory Information Using Deep Learning

A semi-supervised deep learning approach for predicting the functional effects of genomic non-coding variations

Prioritization of Nonsynonymous Single Nucleotide Variants for Exome Sequencing Studies Via Integrative Learning on Multiple Genomic Data

Dissecting molecular regulatory mechanisms underlying noncoding susceptibility SNPs associated with 19 autoimmune diseases using multi-omics integrative analysis

A computational method for identification of disease-associated non-coding SNPs in human genome

Combining Artificial Intelligence: Deep Learning with Hi-C Data to Predict the Functional Effects of Non-Coding Variants.

Integrating sequence conservation features and a domain-domain interaction network to detect disease-associated nsSNPs

Integrating Multiple Genomic Data to Predict Disease-Causing Nonsynonymous Single Nucleotide Variants in Exome Sequencing Studies

Prediction of disease-associated functional variants in noncoding regions through a comprehensive analysis by integrating datasets and features

In silico generation and augmentation of regulatory variants from massively parallel reporter assay using conditional variational autoencoder

Supervised Learning-Based Tagsnp Selection for Genome-Wide Disease Classifications

Dissecting Regulatory Mechanisms Underlying Noncoding GWAS SNPs on 20 Autoimmune Disease

Predicting Functional Elements and Variants Effects in Non-Coding Regions Based on Deep Learning

Identifying noncoding risk variants using disease-relevant gene regulatory networks

Multiomics dissection of molecular regulatory mechanisms underlying autoimmune-associated noncoding SNPs

SDLDA: lncRNA-disease association prediction based on singular value decomposition and deep learning

Predicting the Functional Effects of Human Non-Coding Variants Based on Stacking Ensemble Learning.

DeepnsSNPs: Accurate Prediction of Non-Synonymous Single-Nucleotide Polymorphisms by Combining Multi-Scale Convolutional Neural Network and Residue Environment Information