Dr. Kim was recently awarded a PVAMU RISE Graduate grant (internal) to support a Graduate Research Assistant (GRA). This is a two-year grant to support a GRA at $25,200/year.
Project Title: Machine Learning-Based Identification of Differentially Methylated Regions in Single-Cell RNA Sequencing Data
Principal Investigator: Seungchan Kim, Ph.D.
Project Summary:
Epigenetic modifications, and in particular DNA methylation, are involved in the regulation of genes and the differentiation of cells. The understanding of differentially methylated regions (DMRs) at single-cell resolution can be used to gain decisive insight into disease mechanisms, cancer development, and cellular heterogeneity. On the other hand, conventional bulk methylation analysis misses the observation of cell-type specific methylation dynamics.
This proposal thus aims to create a machine learning (ML) framework for the identification of DMRs in scRNA-seq data. The methodology will evaluate predict DMRs by integrating the gene expression profiles and the methylation DNA sequencing data through the application of deep learning models, such as graph neural networks (GNNs) and transformers.
The model will be trained and validated on publicly available datasets, such as GEO (GSE140493) and ENCODE. ML algorithms in this study enhance the accuracy and resolution of epigenetic landscape analysis, promoting personalized medicine and biomarker development.
Anticipated Results and Impact
This project, therefore, aims to define a single-cell resolution, accurate, machine-learn-based method by integration of multi-omics datasets, thus enhancing biological interpretability, revealing epigenetic regulation mechanisms in cancer, neurodegeneration, and immune diseases. The outcomes could enhance precision medicine, facilitating biomarker discovery for early disease diagnosis and treatment stratification. Additionally, the methodology can be extended to other single-cell multi-omics studies, added to epigenetics, and AI-driven bioinformatics development.