Teng Wang

This study identifies backdoor vulnerabilities in genomic foundation models through data poisoning during fine-tuning and proposes an auditing framework to detect such threats.

The paper shows plasmid genomes have coordinated informational architecture, with tRNAs and other genes organized to meet codon demands and support plasmid function.

The paper identifies a third replicon state in bacteria, intermediate between chromosomes and plasmids, using sequence affinity analysis across multiple genera.

The paper uses machine learning to show that host chromosomal gene content predicts plasmid biology traits like carriage status and mobility potential across prokaryotic genomes.

The paper develops a machine learning model to predict plasmid copy numbers using protein domains and applies it to metagenomic and clinical plasmid data to identify ecological patterns.

This paper introduces genetic devices for precise control of cell-type ratios and multicellular morphology in synthetic biology systems.

LabscriptAI translates natural language into robot scripts for synthetic biology, achieving high simulation accuracy and enabling secure, standardized, and traceable lab automation.

This study shows that horizontal gene transfer promotes alternative stable states in microbial communities through various mechanisms.