Model: openai/gpt-4o

Phenotype Balanced Accuracy Precision Sample Size
Aerophilicity 0.897 0.736 4356
Animal pathogenicity 0.818 0.868 4097
Biofilm formation 0.557 0.736 504
Biosafety level 0.894 0.641 7438
Cell shape 0.785 0.685 6597
Extreme environment tolerance 0.664 0.174 6704
Gram staining 0.811 0.653 3878
Health association 0.567 0.381 23
Hemolysis 0.578 0.443 293
Host association 0.809 0.812 5332
Motility 0.862 0.780 5927
Plant pathogenicity 0.736 0.284 7467
Spore formation 0.932 0.935 7240