Model for clade_thermophilic¶
- Description: Binary classification of a bacterial phenotype (thermophily) from Pfam protein-domain content of the genome, for strains in the BacDive-AI release (13,669 strains x 2,047 Pfam families). Whole gene-content clades are held out, so no test strain shares a clade with any training strain. This is the honest generalisation setting; compare against the bacdive variant to isolate the effect of the split protocol. Note that trait prevalence ranges from 2.6% to 55%, so accuracy is a weak summary; macro-F1 is the informative metric. External links: https://github.com/JKoblitz/bacdive-AI, https://doi.org/10.5281/zenodo.13757323
Reference(s): https://doi.org/10.1038/s42003-025-08182-w, https://github.com/JKoblitz/bacdive-AI;https://doi.org/10.5281/zenodo.13757323, https://doi.org/10.5281/zenodo.13757323, https://github.com/JKoblitz/bacdive-AI
Model benchmarks
| Model name | Dataset | ACC | Team name | Dataset size | Date submitted | Notes |
|---|---|---|---|---|---|---|
| g2p_logistic | bacdive_clade | 0.9875 | JHU | 13210 | 09-07-2026 | CSV, JSON, run.sh, Info |
| g2p_rf | bacdive_clade | 0.9727 | JHU | 13210 | 09-07-2026 | CSV, JSON, run.sh, Info |
| g2p_multitask | bacdive_clade | 0.989 | JHU | 13210 | 09-07-2026 | CSV, JSON, run.sh, Info |
| g2p_gbm | bacdive_clade | 0.9841 | JHU | 13210 | 09-07-2026 | CSV, JSON, run.sh, Info |
| g2p_mlp | bacdive_clade | 0.9879 | JHU | 13210 | 09-07-2026 | CSV, JSON, run.sh, Info |