Data coherence over data volume drives generalisable genome-based prediction of microbial carbon utilisation
Dileep Kishore, Priya Ranjan, Christopher J. Neely, Mikaela Cashman, Zahmeeth Sakkaff, William J. Riehl, Marcin P. Joachimiak, Janaka N. Edirisinghe, José P. Faria, Pamela B. Weisenhorn, Myra B. Cohen, Dale A. Pelletier, Robert W. Cottingham, Christopher S. Henry, Mitchel J. Doktycz, , Paramvir Dehal
This study shows that filtering genomes based on coherent genotype-phenotype relationships improves cross-dataset prediction of microbial carbon utilisation, outperforming larger but less coherent datasets.
bioRxiv · Aug 12, 2026 · PreprintBioinformatics