Xingxu Huang

This paper introduces AMP-SEMiner, an AI framework for discovering antimicrobial peptides from metagenomes, identifying over 1.6 million candidates and validating their antimicrobial activity.

This study benchmarks five CNA inference tools using single-cell multi-omics data, finding Numbat performs best overall while CopyKAT excels with expression matrix input.

The paper introduces CHOOSER, a foundation model-based framework for discovering CRISPR-Cas systems with self-processing pre-crRNA capability, identifying 11 novel homologs and validating one for functional use.

InstructPLM aligns a protein language model with structure instructions to generate diverse proteins, improving performance over existing models in sequence recovery and activity.

The paper reports the structure of AsCas12f1 and engineers it to enhance genome editing activity in human cells.

This paper introduces an iterative sequence refinement pipeline for inverse protein folding that improves sequence recovery and editing activity in transposase variants.

This study uses base editing in goat embryos to introduce nonsense mutations in FGF5, improving cashmere yield with minimal off-target effects.