PROBE Leaderboard: Protein Representation Model Evaluation

Welcome to the PROBE (Protein RepresentatiOn BEnchmark) leaderboard! This platform evaluates protein representation models based on their ability to capture functional properties of proteins through four key benchmarks:

Semantic Similarity Inference
Ontology-based Function Prediction
Drug Target Protein Family Classification
Protein Protein Binding Affinity Estimation

Submit your own representation models and compare their performance across these benchmarks. For more details on how to participate, see the submission guidelines at Submit Here! tab. For descriptions of each benchmark and its metrics, please refer to the About tab. To visualize results, visit the Visualization tab.

If you find PROBE Leaderboard useful, please consider citing our work:

Çevrim, E., Yiğit, M. G., Ulusoy, E., Yılmaz, A., & Doğan, T. (2025). A Benchmarking Platform for Assessing Protein Language Models on Function-related Prediction Tasks. In Protein Function Prediction: Methods and Protocols (pp. 241-268). New York, NY: Springer US.

Unsal, S., Atas, H., Albayrak, M., Turhan, K., Acar, A. C., & Doğan, T. (2022). Learning functional properties of proteins with language models. Nature Machine Intelligence, 4(3), 227-245.


AAC	-0.0121	0.0912	0.2141	0.0978	nan	nan	nan	nan	nan	nan	nan	nan	0.3881	0.4097	0.4867	0.4430	0.1611	0.1872	0.3225	0.1976	0.2394	0.2322	0.2829	0.2819	0.2629	0.2764	0.3640	0.3075	0.6178	0.6013	0.4607	0.5459	0.5508	0.4350	0.5578	0.5505	0.4354	0.5868	0.5675	0.4167	18.4780	105.9310	0.4605
APAAC	0.1717	0.2409	0.2736	0.2287	nan	nan	nan	nan	nan	nan	nan	nan	0.4710	0.5828	0.6744	0.5989	0.2239	0.3447	0.4230	0.4093	0.2396	0.3953	0.4525	0.4603	0.3115	0.4409	0.5166	0.4895	0.4636	0.4722	0.3556	0.2315	0.2661	0.2398	0.3843	0.3888	0.2965	0.3779	0.3854	0.2879	17.9010	102.2670	0.4772
BLAST	0.1127	0.0885	0.3432	0.1815	nan	nan	nan	nan	nan	nan	nan	nan	0.8421	0.8749	0.9219	0.8571	0.4616	0.5559	0.6807	0.5020	0.4851	0.5731	0.6969	0.5298	0.5963	0.6680	0.7665	0.6296	0.8899	0.8837	0.8296	0.8676	0.8537	0.7958	0.8783	0.8678	0.8101	0.7742	0.7458	0.6719	nan	nan	nan
CPC-PROT	0.0605	-0.0870	0.1091	0.0275	nan	nan	nan	nan	nan	nan	nan	nan	0.5183	0.6545	0.7049	0.6819	0.2688	0.3955	0.5084	0.4087	0.3214	0.4352	0.5207	0.4555	0.3695	0.4950	0.5780	0.5153	0.7234	0.7117	0.5852	0.7110	0.6960	0.5716	0.7433	0.7351	0.6058	0.7159	0.7102	0.5676	0.7299	62.6090	0.8309
ESM1B	0.3847	0.3732	0.4152	0.3910	nan	nan	nan	nan	nan	nan	nan	nan	0.7657	0.8324	0.8653	0.8376	0.4152	0.5333	0.6412	0.5463	0.4682	0.6098	0.7013	0.6317	0.5497	0.6585	0.7359	0.6719	0.9336	0.9321	0.8951	0.9359	0.9339	0.8965	0.8978	0.8864	0.8423	0.9151	0.9149	0.8636	0.4828	50.0190	0.8915
ESM2-650m	0.2978	0.2811	0.4274	0.3354	nan	nan	nan	nan	nan	nan	nan	nan	0.8321	0.8743	0.8979	0.8703	0.4671	0.5987	0.6671	0.5977	0.5319	0.6472	0.7147	0.6489	0.6104	0.7067	0.7599	0.7056	0.9328	0.9308	0.8927	0.9397	0.9389	0.9033	0.9300	0.9289	0.8890	0.9163	0.9174	0.8676	0.4759	49.6390	0.8931
ESM3-small	0.5102	0.5762	0.4227	0.5031	nan	nan	nan	nan	nan	nan	nan	nan	0.6839	0.8076	0.8416	0.8261	0.3729	0.4988	0.6074	0.5303	0.3812	0.5176	0.5913	0.5507	0.4793	0.6080	0.6801	0.6357	0.8444	0.8259	0.7794	0.8906	0.8819	0.8260	0.8535	0.8378	0.7926	0.8510	0.8330	0.7775	0.5116	52.3330	0.8888
GENE2VEC^	0.1756	0.3596	0.4080	0.3144	nan	nan	nan	nan	nan	nan	nan	nan	0.4440	0.5311	0.5932	0.5266	0.3116	0.4385	0.4921	0.4214	0.3740	0.5016	0.5499	0.5009	0.3765	0.4904	0.5451	0.4830	0.4978	0.4668	0.2701	0.5044	0.4736	0.2709	0.4887	0.4569	0.2629	0.4540	0.4135	0.2167	nan	nan	nan
HMMER	0.1374	0.3219	0.1577	0.2056	nan	nan	nan	nan	nan	nan	nan	nan	0.8579	0.8908	0.9115	0.8880	0.5022	0.6110	0.6448	0.6037	0.4910	0.5957	0.6451	0.5868	0.6171	0.6992	0.7338	0.6928	0.8886	0.8821	0.8257	0.8786	0.8704	0.8104	0.8847	0.8775	0.8191	0.8048	0.7875	0.7075	nan	nan	nan
K-SEP	0.2222	0.2944	0.2924	0.2697	nan	nan	nan	nan	nan	nan	nan	nan	0.7068	0.8090	0.8480	0.8124	0.3590	0.5169	0.5932	0.5497	0.3445	0.5023	0.5819	0.5385	0.4701	0.6094	0.6744	0.6335	0.7496	0.7277	0.6548	0.7690	0.7695	0.6933	0.8079	0.8001	0.7068	0.7254	0.7100	0.6070	0.9705	72.0730	0.7613
LEARNED-VEC	0.4100	0.3052	0.2987	0.3380	nan	nan	nan	nan	nan	nan	nan	nan	0.5858	0.6755	0.7205	0.6822	0.3066	0.3917	0.4774	0.3870	0.3162	0.4089	0.4695	0.4221	0.4029	0.4920	0.5558	0.4971	0.6879	0.6754	0.5398	0.6814	0.6705	0.5323	0.6921	0.6817	0.5387	0.6564	0.6366	0.4873	11.8480	81.3080	0.7034
MUT2VEC^	0.5462	0.3949	0.5799	0.5070	nan	nan	nan	nan	nan	nan	nan	nan	0.4887	0.5731	0.6264	0.5738	0.3273	0.4271	0.4931	0.4205	0.3604	0.4616	0.5288	0.4561	0.3922	0.4873	0.5495	0.4835	0.5755	0.5542	0.3694	0.5786	0.5583	0.3655	0.5813	0.5618	0.3662	0.5894	0.5724	0.3617	nan	nan	nan
PFAM^	0.3489	0.5107	0.4222	0.4273	nan	nan	nan	nan	nan	nan	nan	nan	0.8148	0.8644	0.9292	0.8366	0.4794	0.5572	0.6724	0.5176	0.5037	0.5806	0.7127	0.5397	0.5993	0.6674	0.7714	0.6313	0.9193	0.9233	0.8750	0.9155	0.9200	0.8696	0.9166	0.9204	0.8701	0.8432	0.8493	0.7610	22.5790	116.9270	0.1917
ProstT5	0.4474	0.4156	0.0806	0.3146	nan	nan	nan	nan	nan	nan	nan	nan	0.8127	0.8745	0.8999	0.8733	0.5450	0.6459	0.7238	0.6259	0.5417	0.6435	0.6991	0.6335	0.6331	0.7213	0.7743	0.7109	0.9317	0.9304	0.8901	0.9280	0.9266	0.8844	0.9266	0.9245	0.8827	0.9098	0.9064	0.8549	0.7844	58.3140	0.8292
ProtALBERT	0.2219	0.3181	0.3690	0.3030	nan	nan	nan	nan	nan	nan	nan	nan	0.8345	0.8854	0.9133	0.8735	0.5106	0.6255	0.7063	0.6011	0.5135	0.6355	0.7256	0.6129	0.6195	0.7155	0.7817	0.6958	0.9419	0.9414	0.9059	0.9395	0.9398	0.9026	0.9364	0.9360	0.8976	0.9033	0.9038	0.8437	0.4183	45.7280	0.9071
ProtBERT-BFD	0.2919	0.4157	0.3235	0.3437	nan	nan	nan	nan	nan	nan	nan	nan	0.7667	0.8482	0.8727	0.8417	0.4867	0.6116	0.6996	0.5894	0.5062	0.6221	0.6941	0.6224	0.5865	0.6939	0.7555	0.6845	0.8790	0.8748	0.8076	0.8762	0.8721	0.8046	0.8805	0.8767	0.8101	0.8561	0.8491	0.7728	0.5736	52.1180	0.8740
ProtT5	0.5664	0.3985	0.2127	0.3925	nan	nan	nan	nan	nan	nan	nan	nan	0.8573	0.9013	0.9224	0.8921	0.5686	0.6558	0.7127	0.6316	0.5884	0.6803	0.7561	0.6616	0.6715	0.7458	0.7970	0.7284	0.9410	0.9399	0.9045	0.9409	0.9396	0.9042	0.9404	0.9388	0.9036	0.9237	0.9221	0.8769	0.6035	5.4620	0.8666
PROTVEC	0.1854	0.2058	0.2953	0.2288	nan	nan	nan	nan	nan	nan	nan	nan	0.5189	0.6367	0.6806	0.6692	0.2383	0.3624	0.4098	0.4215	0.2428	0.3808	0.4526	0.4208	0.3333	0.4600	0.5143	0.5038	0.4383	0.4301	0.3518	0.4565	0.4486	0.3565	0.5457	0.5312	0.4064	0.5372	0.5368	0.4098	11.3410	80.8890	0.7191
ProtXLNET	0.2271	0.2549	0.3090	0.2637	nan	nan	nan	nan	nan	nan	nan	nan	0.7114	0.8158	0.8472	0.8248	0.3729	0.4996	0.6120	0.5095	0.4466	0.5878	0.6673	0.5987	0.5103	0.6344	0.7088	0.6443	0.9040	0.9018	0.8490	0.8677	0.8586	0.7981	0.8511	0.8368	0.7857	0.8411	0.8272	0.7531	0.6068	57.0010	0.8615
SaProt-35m-af2	0.4032	0.3691	0.2419	0.3380	nan	nan	nan	nan	nan	nan	nan	nan	0.7678	0.8310	0.8557	0.8375	0.4591	0.5752	0.6842	0.5486	0.5095	0.6016	0.6860	0.6024	0.5788	0.6693	0.7420	0.6629	0.8955	0.8923	0.8328	0.8904	0.8866	0.8254	0.8982	0.8971	0.8376	0.8758	0.8727	0.8044	nan	nan	nan
SEQVEC	0.4238	0.4242	0.2436	0.3639	nan	nan	nan	nan	nan	nan	nan	nan	0.8366	0.8857	0.9084	0.8779	0.5021	0.6003	0.6470	0.5903	0.5010	0.6065	0.6817	0.5918	0.6132	0.6975	0.7457	0.6867	0.9150	0.9130	0.8630	0.9109	0.9092	0.8572	0.9102	0.9080	0.8557	0.8865	0.8828	0.8175	0.5331	52.4230	0.8801
TAPE-BERT-PFAM^	0.4965	0.2218	0.2068	0.3084	nan	nan	nan	nan	nan	nan	nan	nan	0.7823	0.8466	0.8594	0.8533	0.4048	0.5351	0.6125	0.5465	0.4518	0.5811	0.6603	0.5823	0.5463	0.6543	0.7107	0.6607	0.8329	0.8226	0.7505	0.8393	0.8327	0.7558	0.8223	0.8104	0.7297	0.7868	0.7716	0.6800	nan	nan	nan
TCGA-EMBEDDING	0.0375	0.5033	0.4822	0.3410	nan	nan	nan	nan	nan	nan	nan	nan	0.3026	0.3439	0.4061	0.3573	0.2366	0.3205	0.4167	0.3173	0.3416	0.4119	0.4955	0.4396	0.2936	0.3588	0.4395	0.3714	0.4831	0.4659	0.2476	0.4749	0.4590	0.2444	0.4848	0.4668	0.2493	0.4530	0.4342	0.1985	nan	nan	nan
UNIREP	0.4213	0.3248	0.4695	0.4052	nan	nan	nan	nan	nan	nan	nan	nan	0.7499	0.8232	0.8599	0.8237	0.3760	0.4795	0.5962	0.4915	0.3803	0.5325	0.5883	0.5722	0.5020	0.6117	0.6815	0.6291	0.7739	0.7633	0.6871	0.8253	0.8168	0.7271	0.8064	0.7961	0.7094	0.7034	0.6829	0.6069	0.7321	63.8740	0.8299

PROBE Leaderboard: Protein Representation Model Evaluation

For detailed explanations of the metrics and benchmarks, please refer to the 📝 About tab.

Method-name colours

Metric-cell shading

If a method name ends with ^, it suggests potential suspicions of data leakage related to similarity, function, or family benchmarks.