scGRNom

scGRNom predicts cell-type-specific disease genes and gene regulatory networks by integrating single-cell and multi-omics data to elucidate how genetic variants influence disease phenotypes.

Key Features:

• Integration of Multi-Omics Data: scGRNom integrates single-cell and other multi-omics datasets to predict cell-type-specific disease genes and regulatory networks and to identify key transcription factors and regulatory elements.
• Cell-Type Specificity: The pipeline predicts disease-related genes and regulatory networks for excitatory and inhibitory neurons, microglia, and oligodendrocytes.
• Application to Neurological Disorders: scGRNom has been applied to schizophrenia and Alzheimer's disease to uncover gene regulatory networks relevant to pathogenesis.
• Enrichment Analyses: The workflow performs enrichment analyses to reveal cross-disease and disease-specific functions and pathways at the cell-type level.
• Machine Learning for Clinical Phenotypes: The approach incorporates machine learning to improve predictions of clinical phenotypes by using cell-type disease genes in predictive models.

Scientific Applications:

• Cell-type-resolved disease genetics: Dissecting the genetic basis of complex diseases at the cellular level by linking variants to cell-type-specific regulatory networks.
• Neurobiology research: Investigating molecular mechanisms and gene regulatory networks in neuropsychiatric and neurodegenerative disorders such as schizophrenia and Alzheimer's disease.

Methodology:

Integration of single-cell and multi-omics datasets to construct predictive models of gene regulation focused on transcription factors and regulatory elements, with machine learning applied to predict clinical phenotypes.