Zhi John Lu

BATTER predicts bacterial transcript 3' ends by analyzing conserved stem-loop structures, revealing clade-specific properties and premature termination in AMR genes.

RNAsmol is a sequence-based deep learning framework that predicts RNA-small molecule interactions using data perturbation and augmentation, achieving improved performance over existing methods.

OligoFormer improves siRNA design by integrating thermodynamic calculations, RNA-FM module, and Oligo encoder, outperforming existing methods in efficacy and off-target prediction.

The paper analyzes cell-free multi-omics data from gastrointestinal cancer patients to identify tumor status-informative signatures in plasma.

Pathformer integrates multi-omics data using a pathway-informed Transformer to improve cancer diagnosis and prognosis prediction.

The paper identifies a temporal genetic program in early neurogenesis, revealing molecular heterogeneity and gene expression dynamics in preplate and Cajal-Retzius neurons.

This study shows that epigenetic reprogramming of tissue-specific transcription in cancer cells promotes metastasis, with FOXA2 playing a key role in liver colonization by metastatic colorectal cancer cells.

The paper integrates RNA-seq data from four mammals to map cell states, identifies conserved genes, and finds distinct evolutionary signatures in neural and testis tissues.

The paper identifies recurrently deregulated lncRNAs in HCC and PVTT samples, linking them to metastasis and prognosis using multi-omics data.