Jeremy D. Ratcliff

MicrobeRT is a framework of fine-tuned language models for metagenomic analysis, achieving strong performance in taxonomic classification, antibiotic resistance detection, and pathogenicity assessment.

The paper predicts quaternary structures of viral proteins using AlphaFold2 and AlphaFold-Multimer, identifying new insights into vaccinology and protease targets.

The paper develops a computational model of influenza A virus infection that simulates semi-infectious particles and genome complementation, showing how these factors influence IFN response and viral load in respiratory epithelial cells.

TaxTriage is a metagenomic analysis pipeline that detects putative pathogens in sequencing data using read classification, mapping, and de novo assembly, with comparable performance to cloud-based tools.

This study evaluates parameter-efficient fine-tuning in genomic classification and finds it underperforms compared to full and partial fine-tuning methods.

The paper evaluates a generative model for synthetic bacteriophage genomes, finding they lack realistic compositional biases and can be detected using compositional metrics.

This study compares Oxford Nanopore R9.4.1 and R10.4.1 sequencing chemistries for resolving the hemagglutinin cleavage site in avian influenza viruses, finding improved accuracy with R10.4.1.