Nicholas Keone Lee

The paper uses a convolutional neural network to identify regulatory motifs and their cell type-specific importance from single-cell data, explaining 29% of gene expression variance in mouse tissues.

EvoAug improves genomic DNN generalization and interpretability through evolution-inspired data augmentations while preserving functional integrity.

This study investigates how architectural choices in convolution-based hybrid networks affect motif representations and the interpretability of their attention maps in genomics.