Claire A. Dubin

This study evaluates large language models' ability to generate predictive modeling code for reproductive health using standardized datasets, finding R code generation more successful than Python and some models matching top DREAM challenge teams.

The paper develops a method to infer strain composition in microbial mixed cultures using natural SNPs as barcodes, validated on Coccidioides posadasii isolates and applied to GWAS.

The paper presents a chromosomal-level genome assembly of the Coccidioides posadasii Silveira strain using SMRT sequencing, improving genome annotation and enabling future genomic studies.