--- license: other license_name: cc-by-4.0-and-apache-2.0 license_link: LICENSE pretty_name: Protein concept probing task_categories: - text-classification - token-classification tags: - biology - protein - probing - interpretability configs: - config_name: functional_sites data_files: - split: train path: functional_sites/train.parquet - split: validation path: functional_sites/validation.parquet - split: test path: functional_sites/test.parquet - config_name: go_bp data_files: - split: train path: go_bp/train.parquet - split: validation path: go_bp/validation.parquet - split: test path: go_bp/test.parquet - config_name: go_cc data_files: - split: train path: go_cc/train.parquet - split: validation path: go_cc/validation.parquet - split: test path: go_cc/test.parquet - config_name: go_mf data_files: - split: train path: go_mf/train.parquet - split: validation path: go_mf/validation.parquet - split: test path: go_mf/test.parquet - config_name: interpro_active_site data_files: - split: train path: interpro_active_site/train.parquet - split: validation path: interpro_active_site/validation.parquet - split: test path: interpro_active_site/test.parquet - config_name: interpro_binding_site data_files: - split: train path: interpro_binding_site/train.parquet - split: validation path: interpro_binding_site/validation.parquet - split: test path: interpro_binding_site/test.parquet - config_name: interpro_conserved_site data_files: - split: train path: interpro_conserved_site/train.parquet - split: validation path: interpro_conserved_site/validation.parquet - split: test path: interpro_conserved_site/test.parquet - config_name: interpro_domain data_files: - split: train path: interpro_domain/train.parquet - split: validation path: interpro_domain/validation.parquet - split: test path: interpro_domain/test.parquet - config_name: interpro_family data_files: - split: train path: interpro_family/train.parquet - split: validation path: interpro_family/validation.parquet - split: test path: interpro_family/test.parquet - config_name: interpro_homologous_superfamily data_files: - split: train path: interpro_homologous_superfamily/train.parquet - split: validation path: interpro_homologous_superfamily/validation.parquet - split: test path: interpro_homologous_superfamily/test.parquet - config_name: interpro_repeat data_files: - split: train path: interpro_repeat/train.parquet - split: validation path: interpro_repeat/validation.parquet - split: test path: interpro_repeat/test.parquet - config_name: lipidation data_files: - split: train path: lipidation/train.parquet - split: validation path: lipidation/validation.parquet - split: test path: lipidation/test.parquet - config_name: localization data_files: - split: train path: localization/train.parquet - split: validation path: localization/validation.parquet - split: test path: localization/test.parquet - config_name: metal_ion_binding data_files: - split: train path: metal_ion_binding/train.parquet - split: validation path: metal_ion_binding/validation.parquet - split: test path: metal_ion_binding/test.parquet - config_name: peptide data_files: - split: train path: peptide/train.parquet - split: validation path: peptide/validation.parquet - split: test path: peptide/test.parquet - config_name: phosphorylation data_files: - split: train path: phosphorylation/train.parquet - split: validation path: phosphorylation/validation.parquet - split: test path: phosphorylation/test.parquet - config_name: post_translational_modification data_files: - split: train path: post_translational_modification/train.parquet - split: validation path: post_translational_modification/validation.parquet - split: test path: post_translational_modification/test.parquet - config_name: prot_param data_files: - split: train path: prot_param/train.parquet - split: validation path: prot_param/validation.parquet - split: test path: prot_param/test.parquet - config_name: secondary_structure data_files: - split: train path: secondary_structure/train.parquet - split: validation path: secondary_structure/validation.parquet - split: test path: secondary_structure/test.parquet - config_name: ssp_q3 data_files: - split: train path: ssp_q3/train.parquet - split: validation path: ssp_q3/validation.parquet - split: test path: ssp_q3/test.parquet - config_name: ssp_q8 data_files: - split: train path: ssp_q8/train.parquet - split: validation path: ssp_q8/validation.parquet - split: test path: ssp_q8/test.parquet - config_name: topology data_files: - split: train path: topology/train.parquet - split: validation path: topology/validation.parquet - split: test path: topology/test.parquet --- # Protein concept probing Twenty-two protein annotation tasks for probing what a protein language model's internal states encode, from amino-acid composition to Gene Ontology terms. Thirteen tasks label a whole protein and nine label each residue. The human tasks are reviewed *Homo sapiens* UniProtKB entries shorter than 1,024 residues; four further tasks are xTrimoPGLM benchmark tables. | Config | Level | Task | Proteins | Labels | Train / validation / test | |---|---|---|---|---|---| | `functional_sites` | residue | multiclass | 6,893 | 6 | 4,778 / 1,106 / 1,009 | | `go_bp` | protein | multilabel | 12,255 | 304 | 8,591 / 1,888 / 1,776 | | `go_cc` | protein | multilabel | 16,443 | 164 | 11,409 / 2,457 / 2,577 | | `go_mf` | protein | multilabel | 11,504 | 123 | 8,010 / 1,786 / 1,708 | | `interpro_active_site` | protein | multilabel | 718 | 58 | 482 / 116 / 120 | | `interpro_binding_site` | protein | multilabel | 640 | 39 | 446 / 87 / 107 | | `interpro_conserved_site` | protein | multilabel | 2,948 | 334 | 2,062 / 450 / 436 | | `interpro_domain` | protein | multilabel | 8,003 | 364 | 5,307 / 1,371 / 1,325 | | `interpro_family` | protein | multilabel | 3,601 | 239 | 2,097 / 678 / 826 | | `interpro_homologous_superfamily` | protein | multilabel | 8,198 | 100 | 5,499 / 1,368 / 1,331 | | `interpro_repeat` | protein | multilabel | 1,189 | 65 | 819 / 189 / 181 | | `lipidation` | residue | multiclass | 978 | 25 | 654 / 129 / 195 | | `localization` | protein | multiclass | 7,530 | 10 | 5,321 / 590 / 1,619 | | `metal_ion_binding` | protein | binary | 7,242 | 1 | 5,346 / 599 / 1,297 | | `peptide` | residue | multiclass | 4,977 | 4 | 3,485 / 700 / 792 | | `phosphorylation` | residue | multiclass | 8,648 | 6 | 6,057 / 1,304 / 1,287 | | `post_translational_modification` | residue | multiclass | 15,349 | 12 | 10,722 / 2,400 / 2,227 | | `prot_param` | protein | regression | 18,126 | 31 | 12,747 / 2,777 / 2,602 | | `secondary_structure` | residue | multiclass | 8,986 | 4 | 6,310 / 1,351 / 1,325 | | `ssp_q3` | residue | multiclass | 11,479 | 3 | 9,731 / 1,083 / 665 | | `ssp_q8` | residue | multiclass | 11,479 | 8 | 9,731 / 1,083 / 665 | | `topology` | residue | multiclass | 4,777 | 17 | 3,378 / 754 / 645 | ## Homology leakage The human tasks keep their published cluster splits: MMseqs2 clusters at 20% identity (30% for InterPro active site, binding site, conserved site and repeat) are assigned whole to train plus validation or to test, and validation clusters are drawn at 50% identity from the rest, with 80% coverage of the target sequence. The four benchmark tasks keep their published test split, and validation is a random 10% of their train split, so they hold no homology guarantee between train and validation. ## How it was built 1. Read each curated human table and its recorded train, validation and test assignment. 2. Turn its targets into one `label` column: a multi-hot vector over the InterPro entries or Gene Ontology terms, 31 ProtParam values, or one class index per residue in the contiguous order of the classes the table holds. 3. For Gene Ontology, keep the terms with more than six positive proteins in both validation and test. 4. For the four benchmark tables, split 10% of train off as validation with seed 184502, then drop sequences of 1,024 residues or more. ## Columns | Column | Meaning | |---|---| | `id` | UniProt accession for the human tasks; a stable sequence-derived id for the benchmark tasks | | `sequence` | amino-acid sequence | | `label` | multi-hot list (multilabel), list of 31 floats (regression), integer (binary or multiclass), or one integer per residue (residue tasks) | Each config folder also holds `metadata.json` with the task level, task type and label names. The benchmark tables publish their classes but not the order of their integers, so their label names are `class_`. ## What a model trained on it may say These are probing benchmarks. InterPro entries and many UniProt features are curated with computational evidence, so a probe here measures agreement with the annotations, not experimentally measured function. Residue tasks count unannotated residues as their own class (`-`), which dominates most of them. ## Sources and licence The human tasks are the curated tables of Figshare record 10.6084/m9.figshare.31169506, released under CC BY 4.0, built from UniProtKB (accessed 2025-09-09) and the InterPro entry list (accessed 2025-09-16). The four benchmark tasks (`metal_ion_binding`, `localization`, `ssp_q3`, `ssp_q8`) are from `biomap-research` on Hugging Face under Apache 2.0. ## Files `/train.parquet`, `/validation.parquet`, `/test.parquet` and `/metadata.json` for each of the 22 configs.