Zhenchao Tang

The paper evaluates whether label-free morphology can replace targeted cellular measurements by analyzing paired phase-reporter data across gene knockouts and screens.

scProtoTransformer provides a scalable reference mapping framework across molecular, cellular, and donor levels using a prototype-based Transformer model with improved interpretability and performance.

The paper develops 3DisoGalaxy to study alternative splicing in breast cancer, finding that many transcripts produce translation-supported proteins and that functional divergence is concentrated in specific regions.

This study develops a framework for predicting protein function and domain annotation at isoform resolution using structural and sequence data, showing strong performance in benchmark tests and capturing domain changes across isoforms.

scKAN improves cell-type annotation and marker gene discovery using Kolmogorov-Arnold Networks, achieving a 6.63% higher macro F1 score than existing methods and identifying potential drug repurposing candidates.