Title: scGRNom: a computational pipeline of integrative multi-omics analyses for predicting cell-type disease genes and regulatory networks
Legend: The computational pipeline, scGRNom, for predicting the gene regulatory network (GRN) via multiomics data. The pipeline inputs the chromatin interactions of regulatory elements (e.g., enhancerpromoter), identifies the transcription factor binding sites (TFBSs) on interacting regulatory elements, predicts TF-target gene expression relationships, and finally outputs a gene regulatory network linking TFs (cyan), regulatory elements (purple) to target genes (green). Further, it predicts cell-type disease genes and regulatory elements from GWAS SNPs and cell-type GRNs. With applications to schizophrenia and Alzheimer’s disease, we predicted disease genes and regulatory
networks for excitatory and inhibitory neurons, microglia, and oligodendrocytes. Enrichment analyses revealed cross-disease and disease-specific functions, pathways and protein-protein interactions at the cell-type level.
Citation: Ting Jin*, Peter Rehani*, Mufang Ying*, Jiawei Huang, Shuang Liu, Panos Roussos, Daifeng Wang. (2021). scGRNom: a computational pipeline of integrative multi-omics analyses for predicting cell-type disease genes and regulatory networks, Genome Medicine, 13, 95, 2021.
Abstract: Understanding cell-type-specific gene regulatory mechanisms from genetic variants to diseases remains challenging. To address this, we developed a computational pipeline, scGRNom (single-cell Gene Regulatory Network prediction from multi-omics), to predict cell-type disease genes and regulatory networks including transcription factors and regulatory elements. With applications to schizophrenia and Alzheimer’s disease, we predicted disease genes and regulatory networks for excitatory and inhibitory neurons, microglia, and oligodendrocytes. Further enrichment analyses revealed cross-disease and disease-specific functions and pathways at the cell-type level. Our machine learning analysis also found that cell-type disease genes improved clinical phenotype predictions. scGRNom is a general-purpose tool available
at https://github.com/daifengwanglab/scGRNom.
About the Lab: The Wang lab focuses on the development of interpretable machine learning approaches and bioinformatics tools for understanding the functional genomics, molecular and cellular mechanisms from genotype to phenotype.