WITHDRAWN 1352 Leveraging new genomic LLMs for studying under-annotated microbial genes

Recent advancements in genomic large language models (LLMs) promise novel bioinformatics solutions for microbiome research. Microbial genomic sequences, like natural languages, form a language of life, enabling the adoption of LLMs to extract useful insights from complex microbial ecologies. In this talk, we will first review recent genomic LLMs, emphasizing their application towards metagenomics data. We will then present a particular application, namely, the study of unannotated microbial genes. We will demonstrate, with both bioinformatics evaluations and epidemiological findings based on public data, that novel genomic LLMs can be utilized to congregate otherwise unannotated microbial genes for more powerful downstream analysis and biologically interpretable findings.