Zhe Liu

BaiZe predicts and interprets cellular responses to perturbations using a multi-view framework, showing effectiveness across diverse contexts and improved predictions with ATAC-seq data.

The paper introduces MANDT, a method that extracts consistent embeddings of motor cortical dynamics to improve BCI decoding across sessions and tasks, validated on rhesus monkey data.

This paper introduces RLAgent, an LLM-based framework for RNA-ligand interaction modeling that enables interactive, code-free prediction at the RNA region level.

This paper introduces TAPPI, a deep learning framework that assesses trends in protein-protein interaction variations caused by genetic mutations, achieving state-of-the-art performance and biological relevance.

Memo-Patho predicts pathogenicity of transmembrane protein mutations using an alignment-free deep learning framework with contrastive pretraining, achieving high accuracy on benchmark datasets and manual curation.

This study benchmarks encoding strategies for non-coding SNPs across QTL prediction tasks, emphasizing interactions between encoding, models, and preprocessing.

This paper introduces Methven, a deep learning framework that predicts non-coding mutation effects on DNA methylation at single-cell resolution, showing superior performance in prioritizing functional mutations.

EMO predicts non-coding mutation-induced up- and down-regulation of gene expression using deep learning, achieving high performance across variant distances and cell types.