hydropd / website /src /data /generated /model_leaderboard.json
rikardsaqe's picture
Upload folder using huggingface_hub
47bb217 verified
Raw
History Blame Contribute Delete
77.1 kB
{
"metrics": [
"auroc",
"auprc",
"accuracy",
"mcc",
"f1",
"precision",
"recall",
"specificity",
"brier",
"ece"
],
"metricLabels": {
"auroc": "AUROC",
"auprc": "AUC-PR",
"accuracy": "Accuracy",
"mcc": "MCC",
"f1": "F1",
"precision": "Precision",
"recall": "Sensitivity / recall",
"specificity": "Specificity",
"brier": "Brier",
"ece": "ECE"
},
"nSpecies": 7,
"nDatasets": 28,
"note": "Aggregated across species from 5-fold protein-grouped CV on the 28 usable datasets. Each value is the mean over folds, then over a species' datasets, then across species. Brier and ECE are calibration metrics where lower is better.",
"models": [
{
"id": "xgb__esm2_phys",
"classifier": "xgb",
"featureSet": "esm2_phys",
"label": "XGBoost \u00b7 ESM2 + physicochemical",
"description": "Gradient-boosted decision trees. The most consistent winner across datasets. ESM2 embeddings plus the 14 physicochemical descriptors. Best overall.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8129,
"min": 0.7219,
"max": 0.8914,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7979,
"min": 0.7374,
"max": 0.8811,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7452,
"min": 0.658,
"max": 0.8096,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4486,
"min": 0.2749,
"max": 0.6194,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7588,
"min": 0.6798,
"max": 0.8395,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7423,
"min": 0.6692,
"max": 0.8194,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.801,
"min": 0.705,
"max": 0.8961,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6536,
"min": 0.3823,
"max": 0.7867,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.2058,
"min": 0.1454,
"max": 0.3123,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1639,
"min": 0.0684,
"max": 0.2343,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8914,
"auprc": 0.8811,
"accuracy": 0.8096,
"mcc": 0.6194,
"f1": 0.8161,
"precision": 0.8009,
"recall": 0.8394,
"specificity": 0.7867,
"brier": 0.1454,
"ece": 0.1218
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8721,
"auprc": 0.8556,
"accuracy": 0.7891,
"mcc": 0.5761,
"f1": 0.7895,
"precision": 0.7781,
"recall": 0.804,
"specificity": 0.7759,
"brier": 0.1492,
"ece": 0.0684
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8553,
"auprc": 0.7809,
"accuracy": 0.7588,
"mcc": 0.533,
"f1": 0.7253,
"precision": 0.7045,
"recall": 0.7994,
"specificity": 0.7512,
"brier": 0.1795,
"ece": 0.1552
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8247,
"auprc": 0.8122,
"accuracy": 0.7462,
"mcc": 0.4944,
"f1": 0.7691,
"precision": 0.7208,
"recall": 0.8322,
"specificity": 0.6561,
"brier": 0.1854,
"ece": 0.1304
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7745,
"auprc": 0.7597,
"accuracy": 0.658,
"mcc": 0.3489,
"f1": 0.6923,
"precision": 0.703,
"recall": 0.7307,
"specificity": 0.6084,
"brier": 0.2279,
"ece": 0.2338
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7505,
"auprc": 0.7374,
"accuracy": 0.7929,
"mcc": 0.2749,
"f1": 0.8395,
"precision": 0.8194,
"recall": 0.8961,
"specificity": 0.3823,
"brier": 0.3123,
"ece": 0.2343
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7219,
"auprc": 0.7581,
"accuracy": 0.6619,
"mcc": 0.2934,
"f1": 0.6798,
"precision": 0.6692,
"recall": 0.705,
"specificity": 0.6143,
"brier": 0.2408,
"ece": 0.2031
}
],
"partial": false
},
{
"id": "xgb__pep_phys",
"classifier": "xgb",
"featureSet": "pep_phys",
"label": "XGBoost \u00b7 PepBERT + physicochemical",
"description": "Gradient-boosted decision trees. The most consistent winner across datasets. PepBERT embeddings concatenated with the 14 physicochemical descriptors.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8121,
"min": 0.7295,
"max": 0.8926,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7847,
"min": 0.7179,
"max": 0.8788,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7569,
"min": 0.6844,
"max": 0.8157,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4602,
"min": 0.2287,
"max": 0.6286,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7682,
"min": 0.7062,
"max": 0.8394,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7503,
"min": 0.7047,
"max": 0.8158,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.8083,
"min": 0.7249,
"max": 0.8932,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6689,
"min": 0.3798,
"max": 0.779,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"brier": {
"mean": 0.2008,
"min": 0.1394,
"max": 0.3059,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1562,
"min": 0.0626,
"max": 0.2361,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8926,
"auprc": 0.8788,
"accuracy": 0.8157,
"mcc": 0.6286,
"f1": 0.8249,
"precision": 0.8005,
"recall": 0.8569,
"specificity": 0.7778,
"brier": 0.1394,
"ece": 0.1071
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8735,
"auprc": 0.8571,
"accuracy": 0.7929,
"mcc": 0.5842,
"f1": 0.7936,
"precision": 0.7814,
"recall": 0.8093,
"specificity": 0.779,
"brier": 0.1469,
"ece": 0.0626
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8457,
"auprc": 0.7482,
"accuracy": 0.7626,
"mcc": 0.5449,
"f1": 0.7349,
"precision": 0.7065,
"recall": 0.8187,
"specificity": 0.7451,
"brier": 0.1834,
"ece": 0.1672
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8261,
"auprc": 0.8125,
"accuracy": 0.7499,
"mcc": 0.4996,
"f1": 0.7695,
"precision": 0.7272,
"recall": 0.8239,
"specificity": 0.6711,
"brier": 0.1809,
"ece": 0.1174
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7645,
"auprc": 0.7179,
"accuracy": 0.6844,
"mcc": 0.3668,
"f1": 0.7062,
"precision": 0.716,
"recall": 0.7314,
"specificity": 0.6509,
"brier": 0.2272,
"ece": 0.2209
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7525,
"auprc": 0.7521,
"accuracy": 0.6961,
"mcc": 0.3688,
"f1": 0.709,
"precision": 0.7047,
"recall": 0.7249,
"specificity": 0.6785,
"brier": 0.222,
"ece": 0.1823
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7295,
"auprc": 0.726,
"accuracy": 0.7967,
"mcc": 0.2287,
"f1": 0.8394,
"precision": 0.8158,
"recall": 0.8932,
"specificity": 0.3798,
"brier": 0.3059,
"ece": 0.2361
}
],
"partial": false
},
{
"id": "pfly_ft",
"classifier": "pfly",
"featureSet": null,
"label": "Pfly (fine-tuned)",
"description": "External dlomix Prosit-style detectability model, fine-tuned per dataset on peptide sequence only. Statistically indistinguishable from our XGB/RF/MLP on matched CV. Full metric panel computed on the matched protein-grouped folds.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8077,
"min": 0.6917,
"max": 0.9036,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7756,
"min": 0.6796,
"max": 0.8928,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.706,
"min": 0.5696,
"max": 0.8196,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4361,
"min": 0.2159,
"max": 0.6397,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.6705,
"min": 0.2331,
"max": 0.8285,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"precision": {
"mean": 0.715,
"min": 0.5737,
"max": 0.8142,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"recall": {
"mean": 0.7299,
"min": 0.1549,
"max": 0.9287,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6968,
"min": 0.4033,
"max": 0.9855,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"brier": {
"mean": 0.2185,
"min": 0.1262,
"max": 0.3654,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"ece": {
"mean": 0.1895,
"min": 0.0851,
"max": 0.3843,
"minSpecies": "wheat",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.9036,
"auprc": 0.8928,
"accuracy": 0.8196,
"mcc": 0.6397,
"f1": 0.8285,
"precision": 0.8074,
"recall": 0.8588,
"specificity": 0.7869,
"brier": 0.1262,
"ece": 0.0958
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8588,
"auprc": 0.8392,
"accuracy": 0.788,
"mcc": 0.5773,
"f1": 0.7946,
"precision": 0.7859,
"recall": 0.8124,
"specificity": 0.7658,
"brier": 0.1489,
"ece": 0.0851
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8575,
"auprc": 0.7578,
"accuracy": 0.7675,
"mcc": 0.538,
"f1": 0.7376,
"precision": 0.7075,
"recall": 0.822,
"specificity": 0.7483,
"brier": 0.1644,
"ece": 0.1497
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7956,
"auprc": 0.7828,
"accuracy": 0.7039,
"mcc": 0.4307,
"f1": 0.7015,
"precision": 0.6704,
"recall": 0.7678,
"specificity": 0.6815,
"brier": 0.2058,
"ece": 0.167
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.7951,
"auprc": 0.7746,
"accuracy": 0.5696,
"mcc": 0.2159,
"f1": 0.2331,
"precision": 0.8142,
"recall": 0.1549,
"specificity": 0.9855,
"brier": 0.3654,
"ece": 0.3843
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7517,
"auprc": 0.6796,
"accuracy": 0.6508,
"mcc": 0.387,
"f1": 0.7088,
"precision": 0.5737,
"recall": 0.9287,
"specificity": 0.4033,
"brier": 0.2828,
"ece": 0.2687
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6917,
"auprc": 0.7025,
"accuracy": 0.6428,
"mcc": 0.264,
"f1": 0.6891,
"precision": 0.6458,
"recall": 0.7646,
"specificity": 0.5066,
"brier": 0.236,
"ece": 0.1757
}
],
"partial": false
},
{
"id": "xgb__esm2",
"classifier": "xgb",
"featureSet": "esm2",
"label": "XGBoost \u00b7 ESM2",
"description": "Gradient-boosted decision trees. The most consistent winner across datasets. ESM2-t12-35M protein-language-model embeddings of the peptide sequence.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8065,
"min": 0.6785,
"max": 0.8876,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7945,
"min": 0.7249,
"max": 0.8779,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.743,
"min": 0.6431,
"max": 0.8022,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4453,
"min": 0.2452,
"max": 0.5992,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7578,
"min": 0.6698,
"max": 0.8476,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7404,
"min": 0.6519,
"max": 0.8217,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.8016,
"min": 0.7036,
"max": 0.9112,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6467,
"min": 0.3785,
"max": 0.779,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"brier": {
"mean": 0.2086,
"min": 0.1477,
"max": 0.2995,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1649,
"min": 0.0685,
"max": 0.2382,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Tetradesmus obliquus (microalga)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8876,
"auprc": 0.8779,
"accuracy": 0.7982,
"mcc": 0.5992,
"f1": 0.8062,
"precision": 0.7921,
"recall": 0.8305,
"specificity": 0.7758,
"brier": 0.1477,
"ece": 0.1146
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.87,
"auprc": 0.8535,
"accuracy": 0.7881,
"mcc": 0.5743,
"f1": 0.7876,
"precision": 0.7791,
"recall": 0.7993,
"specificity": 0.779,
"brier": 0.1504,
"ece": 0.0685
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8559,
"auprc": 0.7848,
"accuracy": 0.7595,
"mcc": 0.5403,
"f1": 0.7316,
"precision": 0.7042,
"recall": 0.8162,
"specificity": 0.7473,
"brier": 0.1791,
"ece": 0.1557
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8221,
"auprc": 0.8098,
"accuracy": 0.745,
"mcc": 0.4931,
"f1": 0.7676,
"precision": 0.7208,
"recall": 0.8298,
"specificity": 0.6573,
"brier": 0.1862,
"ece": 0.1316
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7672,
"auprc": 0.7633,
"accuracy": 0.6648,
"mcc": 0.3604,
"f1": 0.6943,
"precision": 0.7133,
"recall": 0.7203,
"specificity": 0.6343,
"brier": 0.235,
"ece": 0.2382
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.764,
"auprc": 0.7473,
"accuracy": 0.8022,
"mcc": 0.3047,
"f1": 0.8476,
"precision": 0.8217,
"recall": 0.9112,
"specificity": 0.3785,
"brier": 0.2995,
"ece": 0.2193
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6785,
"auprc": 0.7249,
"accuracy": 0.6431,
"mcc": 0.2452,
"f1": 0.6698,
"precision": 0.6519,
"recall": 0.7036,
"specificity": 0.555,
"brier": 0.262,
"ece": 0.2267
}
],
"partial": false
},
{
"id": "rf__pep_phys",
"classifier": "rf",
"featureSet": "pep_phys",
"label": "Random forest \u00b7 PepBERT + physicochemical",
"description": "Bagged decision trees. Robust, usually a step behind XGBoost. PepBERT embeddings concatenated with the 14 physicochemical descriptors.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8039,
"min": 0.7356,
"max": 0.881,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.779,
"min": 0.7056,
"max": 0.8685,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7363,
"min": 0.6479,
"max": 0.8013,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.445,
"min": 0.3015,
"max": 0.6,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7532,
"min": 0.6832,
"max": 0.8459,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7379,
"min": 0.6696,
"max": 0.8233,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7998,
"min": 0.7187,
"max": 0.9061,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6479,
"min": 0.3978,
"max": 0.7732,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.194,
"min": 0.1551,
"max": 0.2322,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1706,
"min": 0.111,
"max": 0.2604,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.881,
"auprc": 0.8685,
"accuracy": 0.7993,
"mcc": 0.6,
"f1": 0.8094,
"precision": 0.7938,
"recall": 0.8352,
"specificity": 0.7732,
"brier": 0.1551,
"ece": 0.1395
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8578,
"auprc": 0.8417,
"accuracy": 0.7751,
"mcc": 0.5519,
"f1": 0.7745,
"precision": 0.7705,
"recall": 0.7838,
"specificity": 0.7729,
"brier": 0.1653,
"ece": 0.111
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8536,
"auprc": 0.7725,
"accuracy": 0.7454,
"mcc": 0.5168,
"f1": 0.7182,
"precision": 0.6867,
"recall": 0.8085,
"specificity": 0.723,
"brier": 0.1781,
"ece": 0.185
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8096,
"auprc": 0.8039,
"accuracy": 0.7277,
"mcc": 0.4608,
"f1": 0.7497,
"precision": 0.7113,
"recall": 0.8047,
"specificity": 0.651,
"brier": 0.1879,
"ece": 0.1169
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7466,
"auprc": 0.7056,
"accuracy": 0.6479,
"mcc": 0.3749,
"f1": 0.6913,
"precision": 0.7102,
"recall": 0.7419,
"specificity": 0.6087,
"brier": 0.2227,
"ece": 0.2316
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7434,
"auprc": 0.707,
"accuracy": 0.8013,
"mcc": 0.3091,
"f1": 0.8459,
"precision": 0.8233,
"recall": 0.9061,
"specificity": 0.3978,
"brier": 0.2322,
"ece": 0.2604
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7356,
"auprc": 0.7537,
"accuracy": 0.6574,
"mcc": 0.3015,
"f1": 0.6832,
"precision": 0.6696,
"recall": 0.7187,
"specificity": 0.6085,
"brier": 0.2166,
"ece": 0.1495
}
],
"partial": false
},
{
"id": "mlp__esm2",
"classifier": "mlp",
"featureSet": "esm2",
"label": "Neural net (MLP) \u00b7 ESM2",
"description": "A small multilayer-perceptron neural net over the feature vector. ESM2-t12-35M protein-language-model embeddings of the peptide sequence.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8033,
"min": 0.6771,
"max": 0.8864,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7801,
"min": 0.6785,
"max": 0.8727,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7479,
"min": 0.6328,
"max": 0.821,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4628,
"min": 0.2604,
"max": 0.6083,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7582,
"min": 0.6548,
"max": 0.8592,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7514,
"min": 0.6614,
"max": 0.841,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7888,
"min": 0.6711,
"max": 0.9084,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6799,
"min": 0.4874,
"max": 0.8011,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.2056,
"min": 0.1563,
"max": 0.26,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1742,
"min": 0.1306,
"max": 0.2017,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Tetradesmus obliquus (microalga)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8864,
"auprc": 0.8727,
"accuracy": 0.802,
"mcc": 0.6083,
"f1": 0.8065,
"precision": 0.81,
"recall": 0.8144,
"specificity": 0.8011,
"brier": 0.1563,
"ece": 0.1306
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8687,
"auprc": 0.7812,
"accuracy": 0.7906,
"mcc": 0.5953,
"f1": 0.7636,
"precision": 0.7428,
"recall": 0.8383,
"specificity": 0.7812,
"brier": 0.1703,
"ece": 0.1648
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8633,
"auprc": 0.8426,
"accuracy": 0.7823,
"mcc": 0.5627,
"f1": 0.7786,
"precision": 0.7812,
"recall": 0.7791,
"specificity": 0.7874,
"brier": 0.1831,
"ece": 0.1593
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8171,
"auprc": 0.7979,
"accuracy": 0.7439,
"mcc": 0.4888,
"f1": 0.765,
"precision": 0.7259,
"recall": 0.8171,
"specificity": 0.6682,
"brier": 0.2043,
"ece": 0.175
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7643,
"auprc": 0.7574,
"accuracy": 0.821,
"mcc": 0.409,
"f1": 0.8592,
"precision": 0.841,
"recall": 0.9084,
"specificity": 0.4874,
"brier": 0.26,
"ece": 0.1957
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7465,
"auprc": 0.7306,
"accuracy": 0.6627,
"mcc": 0.3154,
"f1": 0.6799,
"precision": 0.6978,
"recall": 0.6933,
"specificity": 0.6218,
"brier": 0.2167,
"ece": 0.2017
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6771,
"auprc": 0.6785,
"accuracy": 0.6328,
"mcc": 0.2604,
"f1": 0.6548,
"precision": 0.6614,
"recall": 0.6711,
"specificity": 0.6124,
"brier": 0.2486,
"ece": 0.1924
}
],
"partial": false
},
{
"id": "mlp__pep_phys",
"classifier": "mlp",
"featureSet": "pep_phys",
"label": "Neural net (MLP) \u00b7 PepBERT + physicochemical",
"description": "A small multilayer-perceptron neural net over the feature vector. PepBERT embeddings concatenated with the 14 physicochemical descriptors.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8025,
"min": 0.6813,
"max": 0.8898,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7693,
"min": 0.6788,
"max": 0.8767,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7416,
"min": 0.6215,
"max": 0.8153,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4498,
"min": 0.2332,
"max": 0.626,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7509,
"min": 0.6406,
"max": 0.849,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7492,
"min": 0.6546,
"max": 0.8364,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7789,
"min": 0.653,
"max": 0.8861,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6849,
"min": 0.5048,
"max": 0.7977,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.2154,
"min": 0.1486,
"max": 0.2883,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1802,
"min": 0.1304,
"max": 0.2349,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Tetradesmus obliquus (microalga)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8898,
"auprc": 0.8767,
"accuracy": 0.8146,
"mcc": 0.626,
"f1": 0.8189,
"precision": 0.8074,
"recall": 0.8361,
"specificity": 0.7977,
"brier": 0.1486,
"ece": 0.1304
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8664,
"auprc": 0.7754,
"accuracy": 0.7612,
"mcc": 0.5485,
"f1": 0.733,
"precision": 0.7256,
"recall": 0.8035,
"specificity": 0.768,
"brier": 0.1893,
"ece": 0.1892
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8611,
"auprc": 0.8372,
"accuracy": 0.7807,
"mcc": 0.5601,
"f1": 0.7783,
"precision": 0.7763,
"recall": 0.784,
"specificity": 0.7798,
"brier": 0.1831,
"ece": 0.1578
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8203,
"auprc": 0.7996,
"accuracy": 0.7472,
"mcc": 0.4924,
"f1": 0.7615,
"precision": 0.7341,
"recall": 0.7972,
"specificity": 0.6942,
"brier": 0.1977,
"ece": 0.1594
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7651,
"auprc": 0.7322,
"accuracy": 0.8153,
"mcc": 0.3462,
"f1": 0.849,
"precision": 0.8364,
"recall": 0.8861,
"specificity": 0.5048,
"brier": 0.2883,
"ece": 0.1998
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7334,
"auprc": 0.6788,
"accuracy": 0.651,
"mcc": 0.3424,
"f1": 0.6748,
"precision": 0.7103,
"recall": 0.6921,
"specificity": 0.6543,
"brier": 0.2497,
"ece": 0.2349
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6813,
"auprc": 0.685,
"accuracy": 0.6215,
"mcc": 0.2332,
"f1": 0.6406,
"precision": 0.6546,
"recall": 0.653,
"specificity": 0.5956,
"brier": 0.2514,
"ece": 0.1896
}
],
"partial": false
},
{
"id": "rf__esm2_phys",
"classifier": "rf",
"featureSet": "esm2_phys",
"label": "Random forest \u00b7 ESM2 + physicochemical",
"description": "Bagged decision trees. Robust, usually a step behind XGBoost. ESM2 embeddings plus the 14 physicochemical descriptors. Best overall.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8008,
"min": 0.7041,
"max": 0.8793,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.777,
"min": 0.6839,
"max": 0.8644,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.734,
"min": 0.6436,
"max": 0.7998,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.44,
"min": 0.2775,
"max": 0.5934,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7487,
"min": 0.6696,
"max": 0.8486,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.739,
"min": 0.6606,
"max": 0.823,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7891,
"min": 0.7028,
"max": 0.9171,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6496,
"min": 0.3654,
"max": 0.7809,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.1917,
"min": 0.1493,
"max": 0.2403,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1682,
"min": 0.0897,
"max": 0.2958,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8793,
"auprc": 0.8644,
"accuracy": 0.7953,
"mcc": 0.5934,
"f1": 0.8028,
"precision": 0.7936,
"recall": 0.8219,
"specificity": 0.7809,
"brier": 0.1493,
"ece": 0.1254
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8574,
"auprc": 0.8383,
"accuracy": 0.7758,
"mcc": 0.5525,
"f1": 0.7744,
"precision": 0.7729,
"recall": 0.7805,
"specificity": 0.7768,
"brier": 0.1602,
"ece": 0.0897
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8448,
"auprc": 0.7626,
"accuracy": 0.7447,
"mcc": 0.5137,
"f1": 0.7093,
"precision": 0.7035,
"recall": 0.7739,
"specificity": 0.7522,
"brier": 0.1729,
"ece": 0.1762
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8106,
"auprc": 0.8037,
"accuracy": 0.7269,
"mcc": 0.4596,
"f1": 0.7526,
"precision": 0.7065,
"recall": 0.8162,
"specificity": 0.6377,
"brier": 0.1836,
"ece": 0.1071
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7565,
"auprc": 0.7555,
"accuracy": 0.6519,
"mcc": 0.3567,
"f1": 0.6834,
"precision": 0.7126,
"recall": 0.7116,
"specificity": 0.6356,
"brier": 0.2139,
"ece": 0.2293
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7527,
"auprc": 0.6839,
"accuracy": 0.7998,
"mcc": 0.3268,
"f1": 0.8486,
"precision": 0.823,
"recall": 0.9171,
"specificity": 0.3654,
"brier": 0.2403,
"ece": 0.2958
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7041,
"auprc": 0.7305,
"accuracy": 0.6436,
"mcc": 0.2775,
"f1": 0.6696,
"precision": 0.6606,
"recall": 0.7028,
"specificity": 0.5986,
"brier": 0.2219,
"ece": 0.1536
}
],
"partial": false
},
{
"id": "mlp__esm2_phys",
"classifier": "mlp",
"featureSet": "esm2_phys",
"label": "Neural net (MLP) \u00b7 ESM2 + physicochemical",
"description": "A small multilayer-perceptron neural net over the feature vector. ESM2 embeddings plus the 14 physicochemical descriptors. Best overall.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.8002,
"min": 0.6557,
"max": 0.8883,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7782,
"min": 0.6856,
"max": 0.8696,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7412,
"min": 0.6228,
"max": 0.8018,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4421,
"min": 0.2024,
"max": 0.6036,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7535,
"min": 0.6422,
"max": 0.8448,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7444,
"min": 0.6357,
"max": 0.8258,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7871,
"min": 0.6633,
"max": 0.898,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6596,
"min": 0.4202,
"max": 0.8039,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.211,
"min": 0.153,
"max": 0.2808,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
},
"ece": {
"mean": 0.1855,
"min": 0.1241,
"max": 0.2491,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8883,
"auprc": 0.8696,
"accuracy": 0.8012,
"mcc": 0.6036,
"f1": 0.8057,
"precision": 0.8098,
"recall": 0.8099,
"specificity": 0.8039,
"brier": 0.153,
"ece": 0.1241
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8698,
"auprc": 0.7879,
"accuracy": 0.7808,
"mcc": 0.5681,
"f1": 0.7471,
"precision": 0.7302,
"recall": 0.8094,
"specificity": 0.7795,
"brier": 0.1722,
"ece": 0.1561
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8645,
"auprc": 0.8435,
"accuracy": 0.7832,
"mcc": 0.5642,
"f1": 0.7797,
"precision": 0.7812,
"recall": 0.781,
"specificity": 0.7872,
"brier": 0.1804,
"ece": 0.1544
},
{
"species": "wheat",
"n": 5,
"auroc": 0.825,
"auprc": 0.8031,
"accuracy": 0.7526,
"mcc": 0.5065,
"f1": 0.7729,
"precision": 0.7312,
"recall": 0.8273,
"specificity": 0.6748,
"brier": 0.2017,
"ece": 0.175
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7588,
"auprc": 0.7559,
"accuracy": 0.6457,
"mcc": 0.3299,
"f1": 0.682,
"precision": 0.6966,
"recall": 0.721,
"specificity": 0.6002,
"brier": 0.232,
"ece": 0.2226
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7396,
"auprc": 0.7016,
"accuracy": 0.8018,
"mcc": 0.3197,
"f1": 0.8448,
"precision": 0.8258,
"recall": 0.898,
"specificity": 0.4202,
"brier": 0.2569,
"ece": 0.2174
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6557,
"auprc": 0.6856,
"accuracy": 0.6228,
"mcc": 0.2024,
"f1": 0.6422,
"precision": 0.6357,
"recall": 0.6633,
"specificity": 0.5517,
"brier": 0.2808,
"ece": 0.2491
}
],
"partial": false
},
{
"id": "xgb__pepbert",
"classifier": "xgb",
"featureSet": "pepbert",
"label": "XGBoost \u00b7 PepBERT",
"description": "Gradient-boosted decision trees. The most consistent winner across datasets. PepBERT protein-language-model embeddings of the peptide sequence.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7959,
"min": 0.7157,
"max": 0.8745,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7672,
"min": 0.6711,
"max": 0.8616,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7417,
"min": 0.6594,
"max": 0.8,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4408,
"min": 0.2994,
"max": 0.5933,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7547,
"min": 0.6821,
"max": 0.8428,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7393,
"min": 0.6632,
"max": 0.8255,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7942,
"min": 0.7169,
"max": 0.8933,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.659,
"min": 0.4251,
"max": 0.7851,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.2066,
"min": 0.1504,
"max": 0.2837,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.155,
"min": 0.063,
"max": 0.2322,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Tetradesmus obliquus (microalga)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8745,
"auprc": 0.8616,
"accuracy": 0.7961,
"mcc": 0.5933,
"f1": 0.8036,
"precision": 0.7972,
"recall": 0.8186,
"specificity": 0.7851,
"brier": 0.1504,
"ece": 0.1079
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8633,
"auprc": 0.8469,
"accuracy": 0.7821,
"mcc": 0.5625,
"f1": 0.7806,
"precision": 0.7771,
"recall": 0.7871,
"specificity": 0.7799,
"brier": 0.1527,
"ece": 0.063
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8351,
"auprc": 0.7518,
"accuracy": 0.7493,
"mcc": 0.521,
"f1": 0.7229,
"precision": 0.695,
"recall": 0.8075,
"specificity": 0.7324,
"brier": 0.1895,
"ece": 0.166
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8159,
"auprc": 0.8063,
"accuracy": 0.7392,
"mcc": 0.478,
"f1": 0.7588,
"precision": 0.7197,
"recall": 0.8101,
"specificity": 0.664,
"brier": 0.1858,
"ece": 0.1187
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.738,
"auprc": 0.7157,
"accuracy": 0.8,
"mcc": 0.3074,
"f1": 0.8428,
"precision": 0.8255,
"recall": 0.8933,
"specificity": 0.4251,
"brier": 0.2837,
"ece": 0.2006
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7291,
"auprc": 0.6711,
"accuracy": 0.6658,
"mcc": 0.3241,
"f1": 0.6924,
"precision": 0.6974,
"recall": 0.7256,
"specificity": 0.6102,
"brier": 0.2442,
"ece": 0.2322
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7157,
"auprc": 0.7169,
"accuracy": 0.6594,
"mcc": 0.2994,
"f1": 0.6821,
"precision": 0.6632,
"recall": 0.7169,
"specificity": 0.6161,
"brier": 0.2401,
"ece": 0.1968
}
],
"partial": false
},
{
"id": "mlp__pepbert",
"classifier": "mlp",
"featureSet": "pepbert",
"label": "Neural net (MLP) \u00b7 PepBERT",
"description": "A small multilayer-perceptron neural net over the feature vector. PepBERT protein-language-model embeddings of the peptide sequence.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7943,
"min": 0.6962,
"max": 0.8779,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7669,
"min": 0.6724,
"max": 0.8611,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7398,
"min": 0.6336,
"max": 0.8212,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4353,
"min": 0.2595,
"max": 0.5983,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7503,
"min": 0.6468,
"max": 0.8539,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7389,
"min": 0.6553,
"max": 0.8326,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7851,
"min": 0.6564,
"max": 0.9005,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6641,
"min": 0.477,
"max": 0.7734,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"brier": {
"mean": 0.2179,
"min": 0.1579,
"max": 0.3056,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1791,
"min": 0.1331,
"max": 0.2147,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8779,
"auprc": 0.8611,
"accuracy": 0.7991,
"mcc": 0.5983,
"f1": 0.8084,
"precision": 0.7936,
"recall": 0.8329,
"specificity": 0.7732,
"brier": 0.1579,
"ece": 0.1331
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8585,
"auprc": 0.8349,
"accuracy": 0.7793,
"mcc": 0.5568,
"f1": 0.7783,
"precision": 0.7728,
"recall": 0.7872,
"specificity": 0.7734,
"brier": 0.1857,
"ece": 0.1601
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8508,
"auprc": 0.7561,
"accuracy": 0.7531,
"mcc": 0.5229,
"f1": 0.7185,
"precision": 0.7165,
"recall": 0.7735,
"specificity": 0.7616,
"brier": 0.1854,
"ece": 0.179
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8096,
"auprc": 0.7942,
"accuracy": 0.7379,
"mcc": 0.4758,
"f1": 0.7544,
"precision": 0.7211,
"recall": 0.7991,
"specificity": 0.6735,
"brier": 0.2075,
"ece": 0.1708
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.757,
"auprc": 0.7352,
"accuracy": 0.8212,
"mcc": 0.3338,
"f1": 0.8539,
"precision": 0.8326,
"recall": 0.9005,
"specificity": 0.477,
"brier": 0.3056,
"ece": 0.2147
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7104,
"auprc": 0.6724,
"accuracy": 0.6547,
"mcc": 0.2998,
"f1": 0.6919,
"precision": 0.6807,
"recall": 0.746,
"specificity": 0.5542,
"brier": 0.2401,
"ece": 0.207
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6962,
"auprc": 0.7146,
"accuracy": 0.6336,
"mcc": 0.2595,
"f1": 0.6468,
"precision": 0.6553,
"recall": 0.6564,
"specificity": 0.6355,
"brier": 0.2429,
"ece": 0.189
}
],
"partial": false
},
{
"id": "rf__esm2",
"classifier": "rf",
"featureSet": "esm2",
"label": "Random forest \u00b7 ESM2",
"description": "Bagged decision trees. Robust, usually a step behind XGBoost. ESM2-t12-35M protein-language-model embeddings of the peptide sequence.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7937,
"min": 0.6788,
"max": 0.8775,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7765,
"min": 0.7027,
"max": 0.8636,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7235,
"min": 0.6115,
"max": 0.7918,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4265,
"min": 0.2134,
"max": 0.5883,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7413,
"min": 0.6489,
"max": 0.8425,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.733,
"min": 0.6324,
"max": 0.8229,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7839,
"min": 0.6863,
"max": 0.9099,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.637,
"min": 0.3632,
"max": 0.7851,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.1941,
"min": 0.1529,
"max": 0.2387,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1656,
"min": 0.0924,
"max": 0.29,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8775,
"auprc": 0.8636,
"accuracy": 0.7918,
"mcc": 0.5883,
"f1": 0.7992,
"precision": 0.7952,
"recall": 0.8139,
"specificity": 0.7851,
"brier": 0.1529,
"ece": 0.126
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.855,
"auprc": 0.8359,
"accuracy": 0.7727,
"mcc": 0.5463,
"f1": 0.7701,
"precision": 0.7724,
"recall": 0.7725,
"specificity": 0.7786,
"brier": 0.1621,
"ece": 0.0924
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8434,
"auprc": 0.7655,
"accuracy": 0.7458,
"mcc": 0.5206,
"f1": 0.7155,
"precision": 0.7011,
"recall": 0.7971,
"specificity": 0.7447,
"brier": 0.1734,
"ece": 0.1699
},
{
"species": "wheat",
"n": 5,
"auroc": 0.8073,
"auprc": 0.801,
"accuracy": 0.7255,
"mcc": 0.4562,
"f1": 0.7509,
"precision": 0.7056,
"recall": 0.8134,
"specificity": 0.6371,
"brier": 0.1854,
"ece": 0.1087
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7519,
"auprc": 0.7027,
"accuracy": 0.7901,
"mcc": 0.3379,
"f1": 0.8425,
"precision": 0.8229,
"recall": 0.9099,
"specificity": 0.3632,
"brier": 0.2387,
"ece": 0.29
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7419,
"auprc": 0.7514,
"accuracy": 0.6271,
"mcc": 0.3225,
"f1": 0.6622,
"precision": 0.7013,
"recall": 0.6863,
"specificity": 0.6165,
"brier": 0.2177,
"ece": 0.2235
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6788,
"auprc": 0.7156,
"accuracy": 0.6115,
"mcc": 0.2134,
"f1": 0.6489,
"precision": 0.6324,
"recall": 0.6944,
"specificity": 0.5341,
"brier": 0.2282,
"ece": 0.1484
}
],
"partial": false
},
{
"id": "xgb__phys",
"classifier": "xgb",
"featureSet": "phys",
"label": "XGBoost \u00b7 Physicochemical",
"description": "Gradient-boosted decision trees. The most consistent winner across datasets. 14 physicochemical descriptors only (length, charge, hydrophobicity, \u2026). Fast, no embeddings.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7889,
"min": 0.738,
"max": 0.8668,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7618,
"min": 0.6771,
"max": 0.8491,
"minSpecies": "Soybean (Glycine max)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7027,
"min": 0.6264,
"max": 0.7875,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4165,
"min": 0.3067,
"max": 0.5749,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7117,
"min": 0.649,
"max": 0.7966,
"minSpecies": "Soybean (Glycine max)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"precision": {
"mean": 0.7043,
"min": 0.6333,
"max": 0.7895,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"recall": {
"mean": 0.7422,
"min": 0.6912,
"max": 0.8119,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6616,
"min": 0.4654,
"max": 0.7723,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.2149,
"min": 0.1622,
"max": 0.2858,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1894,
"min": 0.0653,
"max": 0.3947,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8668,
"auprc": 0.8491,
"accuracy": 0.7875,
"mcc": 0.5749,
"f1": 0.7966,
"precision": 0.7895,
"recall": 0.8119,
"specificity": 0.7723,
"brier": 0.1622,
"ece": 0.1309
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8375,
"auprc": 0.8091,
"accuracy": 0.7597,
"mcc": 0.5175,
"f1": 0.7616,
"precision": 0.7475,
"recall": 0.7794,
"specificity": 0.7417,
"brier": 0.1657,
"ece": 0.0653
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7874,
"auprc": 0.76,
"accuracy": 0.6807,
"mcc": 0.3765,
"f1": 0.7017,
"precision": 0.7197,
"recall": 0.7183,
"specificity": 0.6611,
"brier": 0.2189,
"ece": 0.2164
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.781,
"auprc": 0.6771,
"accuracy": 0.6789,
"mcc": 0.396,
"f1": 0.649,
"precision": 0.6453,
"recall": 0.7208,
"specificity": 0.6825,
"brier": 0.2296,
"ece": 0.2106
},
{
"species": "wheat",
"n": 5,
"auroc": 0.7677,
"auprc": 0.7521,
"accuracy": 0.699,
"mcc": 0.3968,
"f1": 0.7217,
"precision": 0.6847,
"recall": 0.7696,
"specificity": 0.6256,
"brier": 0.2107,
"ece": 0.1344
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7436,
"auprc": 0.7387,
"accuracy": 0.6264,
"mcc": 0.3067,
"f1": 0.6585,
"precision": 0.6333,
"recall": 0.7042,
"specificity": 0.4654,
"brier": 0.2858,
"ece": 0.3947
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.738,
"auprc": 0.7466,
"accuracy": 0.6867,
"mcc": 0.3471,
"f1": 0.693,
"precision": 0.7098,
"recall": 0.6912,
"specificity": 0.6826,
"brier": 0.2316,
"ece": 0.1735
}
],
"partial": false
},
{
"id": "rf__phys",
"classifier": "rf",
"featureSet": "phys",
"label": "Random forest \u00b7 Physicochemical",
"description": "Bagged decision trees. Robust, usually a step behind XGBoost. 14 physicochemical descriptors only (length, charge, hydrophobicity, \u2026). Fast, no embeddings.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7886,
"min": 0.7232,
"max": 0.8721,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7625,
"min": 0.6942,
"max": 0.8531,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.6994,
"min": 0.6061,
"max": 0.7916,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4121,
"min": 0.2529,
"max": 0.5853,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7084,
"min": 0.6435,
"max": 0.8018,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"precision": {
"mean": 0.7045,
"min": 0.6196,
"max": 0.7896,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"recall": {
"mean": 0.7383,
"min": 0.6942,
"max": 0.8239,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"specificity": {
"mean": 0.6582,
"min": 0.4139,
"max": 0.7706,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.1931,
"min": 0.1472,
"max": 0.2444,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"ece": {
"mean": 0.1657,
"min": 0.0619,
"max": 0.4021,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8721,
"auprc": 0.8531,
"accuracy": 0.7916,
"mcc": 0.5853,
"f1": 0.8018,
"precision": 0.7896,
"recall": 0.8239,
"specificity": 0.7706,
"brier": 0.1472,
"ece": 0.0989
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8377,
"auprc": 0.8141,
"accuracy": 0.7573,
"mcc": 0.5145,
"f1": 0.7579,
"precision": 0.75,
"recall": 0.7701,
"specificity": 0.7487,
"brier": 0.1648,
"ece": 0.0619
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.7958,
"auprc": 0.6945,
"accuracy": 0.6905,
"mcc": 0.4078,
"f1": 0.646,
"precision": 0.659,
"recall": 0.697,
"specificity": 0.7169,
"brier": 0.1924,
"ece": 0.1645
},
{
"species": "wheat",
"n": 5,
"auroc": 0.7723,
"auprc": 0.7612,
"accuracy": 0.7005,
"mcc": 0.402,
"f1": 0.7237,
"precision": 0.6861,
"recall": 0.7745,
"specificity": 0.6242,
"brier": 0.1934,
"ece": 0.0886
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7714,
"auprc": 0.759,
"accuracy": 0.6775,
"mcc": 0.3885,
"f1": 0.6949,
"precision": 0.7324,
"recall": 0.6974,
"specificity": 0.6867,
"brier": 0.2025,
"ece": 0.207
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7478,
"auprc": 0.7614,
"accuracy": 0.6723,
"mcc": 0.3338,
"f1": 0.691,
"precision": 0.695,
"recall": 0.7108,
"specificity": 0.6465,
"brier": 0.2069,
"ece": 0.1367
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7232,
"auprc": 0.6942,
"accuracy": 0.6061,
"mcc": 0.2529,
"f1": 0.6435,
"precision": 0.6196,
"recall": 0.6942,
"specificity": 0.4139,
"brier": 0.2444,
"ece": 0.4021
}
],
"partial": false
},
{
"id": "rf__pepbert",
"classifier": "rf",
"featureSet": "pepbert",
"label": "Random forest \u00b7 PepBERT",
"description": "Bagged decision trees. Robust, usually a step behind XGBoost. PepBERT protein-language-model embeddings of the peptide sequence.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7854,
"min": 0.7005,
"max": 0.8648,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7622,
"min": 0.6737,
"max": 0.8544,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7172,
"min": 0.5973,
"max": 0.7993,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4168,
"min": 0.2846,
"max": 0.5584,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7354,
"min": 0.6469,
"max": 0.8441,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7285,
"min": 0.6583,
"max": 0.8262,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7781,
"min": 0.6966,
"max": 0.8994,
"minSpecies": "Tetradesmus obliquus (microalga)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.641,
"min": 0.4167,
"max": 0.7791,
"minSpecies": "Faba bean (Vicia faba)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"brier": {
"mean": 0.2021,
"min": 0.1722,
"max": 0.2332,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Tetradesmus obliquus (microalga)",
"nSpecies": 7
},
"ece": {
"mean": 0.1782,
"min": 0.1194,
"max": 0.2816,
"minSpecies": "wheat",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8648,
"auprc": 0.8544,
"accuracy": 0.775,
"mcc": 0.5584,
"f1": 0.7834,
"precision": 0.7843,
"recall": 0.7963,
"specificity": 0.7724,
"brier": 0.1722,
"ece": 0.1562
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8473,
"auprc": 0.8327,
"accuracy": 0.764,
"mcc": 0.5314,
"f1": 0.7599,
"precision": 0.7692,
"recall": 0.7571,
"specificity": 0.7791,
"brier": 0.1757,
"ece": 0.1261
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.8392,
"auprc": 0.763,
"accuracy": 0.7299,
"mcc": 0.4861,
"f1": 0.7028,
"precision": 0.6751,
"recall": 0.7892,
"specificity": 0.71,
"brier": 0.1823,
"ece": 0.1758
},
{
"species": "wheat",
"n": 5,
"auroc": 0.7952,
"auprc": 0.7892,
"accuracy": 0.717,
"mcc": 0.4398,
"f1": 0.7402,
"precision": 0.7024,
"recall": 0.7949,
"specificity": 0.6406,
"brier": 0.195,
"ece": 0.1194
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7334,
"auprc": 0.6973,
"accuracy": 0.7993,
"mcc": 0.3213,
"f1": 0.8441,
"precision": 0.8262,
"recall": 0.8994,
"specificity": 0.4167,
"brier": 0.2321,
"ece": 0.2816
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.7175,
"auprc": 0.7251,
"accuracy": 0.6382,
"mcc": 0.2846,
"f1": 0.6707,
"precision": 0.6583,
"recall": 0.713,
"specificity": 0.6007,
"brier": 0.2241,
"ece": 0.1555
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7005,
"auprc": 0.6737,
"accuracy": 0.5973,
"mcc": 0.2963,
"f1": 0.6469,
"precision": 0.6841,
"recall": 0.6966,
"specificity": 0.5675,
"brier": 0.2332,
"ece": 0.2331
}
],
"partial": false
},
{
"id": "mlp__phys",
"classifier": "mlp",
"featureSet": "phys",
"label": "Neural net (MLP) \u00b7 Physicochemical",
"description": "A small multilayer-perceptron neural net over the feature vector. 14 physicochemical descriptors only (length, charge, hydrophobicity, \u2026). Fast, no embeddings.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.7643,
"min": 0.6271,
"max": 0.8671,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"auprc": {
"mean": 0.7243,
"min": 0.6339,
"max": 0.8428,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.7157,
"min": 0.5903,
"max": 0.8229,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"mcc": {
"mean": 0.4052,
"min": 0.1912,
"max": 0.5692,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"f1": {
"mean": 0.7286,
"min": 0.6275,
"max": 0.8544,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"precision": {
"mean": 0.7211,
"min": 0.6293,
"max": 0.8444,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"recall": {
"mean": 0.7657,
"min": 0.6688,
"max": 0.8887,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
},
"specificity": {
"mean": 0.647,
"min": 0.5337,
"max": 0.7621,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Spirulina (Arthrospira platensis)",
"nSpecies": 7
},
"brier": {
"mean": 0.2132,
"min": 0.1581,
"max": 0.2483,
"minSpecies": "Spirulina (Arthrospira platensis)",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
},
"ece": {
"mean": 0.1778,
"min": 0.1183,
"max": 0.2741,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Faba bean (Vicia faba)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.8671,
"auprc": 0.8428,
"accuracy": 0.7817,
"mcc": 0.5692,
"f1": 0.7919,
"precision": 0.7815,
"recall": 0.8151,
"specificity": 0.7621,
"brier": 0.1581,
"ece": 0.1185
},
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.8255,
"auprc": 0.7912,
"accuracy": 0.747,
"mcc": 0.493,
"f1": 0.7467,
"precision": 0.7403,
"recall": 0.7579,
"specificity": 0.7378,
"brier": 0.1863,
"ece": 0.1183
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.7912,
"auprc": 0.6992,
"accuracy": 0.7027,
"mcc": 0.4322,
"f1": 0.6638,
"precision": 0.6707,
"recall": 0.7191,
"specificity": 0.73,
"brier": 0.2327,
"ece": 0.2195
},
{
"species": "wheat",
"n": 5,
"auroc": 0.7551,
"auprc": 0.7325,
"accuracy": 0.693,
"mcc": 0.3873,
"f1": 0.7156,
"precision": 0.6822,
"recall": 0.7612,
"specificity": 0.6242,
"brier": 0.2193,
"ece": 0.1475
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.7503,
"auprc": 0.6908,
"accuracy": 0.6721,
"mcc": 0.3758,
"f1": 0.7004,
"precision": 0.6996,
"recall": 0.7491,
"specificity": 0.6011,
"brier": 0.2119,
"ece": 0.1918
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.7341,
"auprc": 0.6794,
"accuracy": 0.8229,
"mcc": 0.3874,
"f1": 0.8544,
"precision": 0.8444,
"recall": 0.8887,
"specificity": 0.54,
"brier": 0.2357,
"ece": 0.2741
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.6271,
"auprc": 0.6339,
"accuracy": 0.5903,
"mcc": 0.1912,
"f1": 0.6275,
"precision": 0.6293,
"recall": 0.6688,
"specificity": 0.5337,
"brier": 0.2483,
"ece": 0.1746
}
],
"partial": false
},
{
"id": "pfly_oob",
"classifier": "pfly",
"featureSet": null,
"label": "Pfly (out-of-the-box)",
"description": "The same external dlomix detectability model used off-the-shelf, with no per-dataset fine-tuning. Significantly worse than the fine-tuned variant and our models on matched CV.",
"nSpecies": 7,
"nDatasets": 28,
"agg": {
"auroc": {
"mean": 0.6311,
"min": 0.5927,
"max": 0.7084,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"auprc": {
"mean": 0.6277,
"min": 0.5616,
"max": 0.6796,
"minSpecies": "Soybean (Glycine max)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"accuracy": {
"mean": 0.5686,
"min": 0.4811,
"max": 0.6447,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"mcc": {
"mean": 0.1585,
"min": 0.0844,
"max": 0.2962,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "Potato (Solanum tuberosum)",
"nSpecies": 7
},
"f1": {
"mean": 0.4638,
"min": 0.1869,
"max": 0.6733,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "wheat",
"nSpecies": 7
},
"precision": {
"mean": 0.605,
"min": 0.5086,
"max": 0.6909,
"minSpecies": "Soybean (Glycine max)",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
},
"recall": {
"mean": 0.4448,
"min": 0.1103,
"max": 0.7868,
"minSpecies": "Red rice bran (Oryza sativa)",
"maxSpecies": "wheat",
"nSpecies": 7
},
"specificity": {
"mean": 0.6978,
"min": 0.4366,
"max": 0.935,
"minSpecies": "wheat",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
},
"brier": {
"mean": 0.3247,
"min": 0.261,
"max": 0.4302,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
},
"ece": {
"mean": 0.2903,
"min": 0.1948,
"max": 0.4326,
"minSpecies": "Potato (Solanum tuberosum)",
"maxSpecies": "Red rice bran (Oryza sativa)",
"nSpecies": 7
}
},
"bySpecies": [
{
"species": "Potato (Solanum tuberosum)",
"n": 16,
"auroc": 0.7084,
"auprc": 0.6796,
"accuracy": 0.6447,
"mcc": 0.2962,
"f1": 0.6439,
"precision": 0.6431,
"recall": 0.6595,
"specificity": 0.6341,
"brier": 0.261,
"ece": 0.1948
},
{
"species": "wheat",
"n": 5,
"auroc": 0.6619,
"auprc": 0.6642,
"accuracy": 0.6174,
"mcc": 0.2461,
"f1": 0.6733,
"precision": 0.5959,
"recall": 0.7868,
"specificity": 0.4366,
"brier": 0.2816,
"ece": 0.229
},
{
"species": "Spirulina (Arthrospira platensis)",
"n": 3,
"auroc": 0.6476,
"auprc": 0.6317,
"accuracy": 0.5193,
"mcc": 0.0986,
"f1": 0.2909,
"precision": 0.6285,
"recall": 0.2129,
"specificity": 0.8663,
"brier": 0.3649,
"ece": 0.3456
},
{
"species": "Tetradesmus obliquus (microalga)",
"n": 1,
"auroc": 0.6081,
"auprc": 0.5801,
"accuracy": 0.568,
"mcc": 0.1126,
"f1": 0.4721,
"precision": 0.5495,
"recall": 0.42,
"specificity": 0.6942,
"brier": 0.3059,
"ece": 0.2688
},
{
"species": "Soybean (Glycine max)",
"n": 1,
"auroc": 0.6023,
"auprc": 0.5616,
"accuracy": 0.5523,
"mcc": 0.0902,
"f1": 0.5108,
"precision": 0.5086,
"recall": 0.5465,
"specificity": 0.5364,
"brier": 0.3257,
"ece": 0.3082
},
{
"species": "Faba bean (Vicia faba)",
"n": 1,
"auroc": 0.5969,
"auprc": 0.6404,
"accuracy": 0.5977,
"mcc": 0.1814,
"f1": 0.4689,
"precision": 0.6182,
"recall": 0.3779,
"specificity": 0.782,
"brier": 0.3039,
"ece": 0.2531
},
{
"species": "Red rice bran (Oryza sativa)",
"n": 1,
"auroc": 0.5927,
"auprc": 0.6364,
"accuracy": 0.4811,
"mcc": 0.0844,
"f1": 0.1869,
"precision": 0.6909,
"recall": 0.1103,
"specificity": 0.935,
"brier": 0.4302,
"ece": 0.4326
}
],
"partial": false
}
]
}