Deep learning in single-cell genomics

Application of deep learning to genomic studies is an exciting area that is rapidly evolving and holds the potential to revolutionize genomics. Our group employs deep learning techniques to tackle several problems related to single-cell RNA-seq data analysis.

• Pseudotime and trajectory analysis
• Prediction of circadian phase
• Single-cell clustering and batch effect removal
• Transfer learning for single-cell clustering
• Joint gene expression imputation, clustering and batch effect removal