Pith. sign in

REVIEW 3 major objections 4 minor 33 references

LC-SEPLM: long-range contact-supervised adaptation for sequence-only protein representation learning

T0 review · 3 major / 4 minor · reviewed 2026-08-01 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Long-range contact supervision during adaptation, removed before inference, improves all eight protein-level tasks of a sequence-only protein language model, with the largest gain in remote-homology recognition.

desk verdict A genuinely new contact-supervision adaptation for sequence-only protein LMs, with a large remote-homology gain that looks real but is currently unverifiable due to an unresolved filtering ambiguity and overreaching all-eight-task claim. read the letter →

arxiv 2607.22777 v2 pith:B7T257DQ submitted 2026-07-24 cs.LG cs.AI

Chen Wang , Boming Kang , Qinghua Cui This is my paper · ORCID
classification cs.LGcs.AI
keywords proteinlanguagemodelsresidue-contactsupervisionlow-rankadaptationremote-homologyrecognitionsequence-onlyinferenceESM2AlphaFoldSwiss-Prottransferlearning
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

LC-SEPLM asks whether a sequence-only protein language model can be improved by supervising it, during adaptation only, with which residue pairs that are far apart along the chain end up close in three-dimensional space. The paper reports that this long-range contact supervision, applied to ESM2 through low-rank adapters, improves all eight protein-level benchmark tasks relative to the original model. The clearest gain is remote-homology recognition, where macro-F1 rises from about 0.61 to 0.68. The benefit is bounded: mutation-effect prediction and post-translational-modification site prediction do not improve, and the model does not beat a structure-aware competitor on every ESM-S benchmark split. If correct, the result shows a practical middle path between sequence-only and structure-requiring protein representations.

What carries the argument

The central object is the pair-specific query contact head. For a sampled pair (i,j), the model forms a descriptor from the two ESM2 residue states, their absolute difference, their element-wise product, and normalized sequence distance; a small MLP projects this descriptor into a query that attends over every residue in the sequence via cross-attention. The attended vector is classified as contact or non-contact. This head is trained together with an auxiliary MLM loss (weight 0.6) while LoRA adapters (rank 4) modify the last eight layers of ESM2-150M. After adaptation the head and structural labels are discarded, leaving a sequence-only encoder.

What would settle it

Rerun the eight controlled tasks on the strictly 30%-identity filtered evaluation sets described in Figure 6 and compare ESM2 against LC-SEPLM; if the remote-homology gain shrinks toward zero once adaptation-set homologs are removed, the contact-supervision effect is not established. A second check: recompute the ESM-S EC gain with the same filtering rule applied to the adaptation set and the benchmark test proteins.

Watch

Extended reading notes

Core claim

The paper introduces a pair-query contact head that, for each sampled residue pair, combines the two residue representations with their difference, element-wise product, and normalized sequence distance, then lets the resulting query attend to the full sequence before predicting whether the pair forms a long-range contact. Positive contacts are defined by sequence separation at least 12 residues and C-alpha distance at most 8 Å, using AlphaFold-derived Swiss-Prot structures as labels. The contact head and an auxiliary masked-language-modeling head are used only during adaptation; after training, they are removed and downstream inference uses the adapted ESM2 encoder on amino-acid sequences a

Load-bearing premise

The main claim assumes that the downstream comparisons are not inflated by overlap between the 500,000 Swiss-Prot adaptation proteins and the benchmark test sets; the manuscript reports identity-filtering audits showing only 1.7-9.0% of evaluation entries remain after strict filtering, yet it never states which filtered split produced Table 2.

Editorial extensions

If this is right

  • Remote-homology recognition can be improved substantially by contact-supervised adaptation without requiring structures at test time.
  • A masked-language-modeling objective can be retained during structural adaptation without collapsing sequence modeling accuracy.
  • Protein-level tasks that depend on global fold organization are the ones that benefit; local diagnostics such as variant effect and PTM-site prediction do not.
  • Checkpoint selection based on downstream validation, rather than raw contact accuracy, is necessary because the objective that peaks upstream is not the one that transfers best.
  • Contact-supervised adaptation can be combined with LoRA, so the added parameters remain in the encoder and no structural branch is needed during deployment.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If the reported gains survive strict sequence-identity filtering, contact supervision may be a lighter-weight alternative to structure-token models for sequence-only annotation pipelines.
  • The method could be tested on larger base encoders; if the remote-homology gain scales, contact supervision may complement rather than replace evolutionary-scale pretraining.
  • Confidence-aware weighting of predicted contacts, or use of experimental structures, could separate genuine transferable signal from AlphaFold prediction bias.
  • The absence of improvement on residue-level tasks suggests a natural test: add a residue-pair objective that supervises local contacts or variant effects to see whether the global/local boundary is due to supervision granularity rather than the encoder.
Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

3 major / 4 minor

Summary. The paper introduces LC-SEPLM, a method that adapts ESM2-150M using LoRA and binary long-range contact labels (|i-j|>=12, Cα distance <= 8 Å) derived from AlphaFold Swiss-Prot structures, with an auxiliary MLM loss. During downstream inference the contact head is removed and the adapted encoder is used sequence-only. The authors report that LC-SEPLM improves all eight protein-level tasks relative to ESM2 at the downstream-selected epoch-10 checkpoint, with the largest gain in Pfam-clan remote homology (macro-F1 0.6122 → 0.6769). They also report an independent comparison with ESM-S on the official benchmark, with gains on EC splits and mixed results elsewhere. Two local diagnostics (ProteinGym and PTM-site prediction) do not improve.

Significance. If the empirical claims are reliable, the paper makes a useful contribution: it demonstrates a bounded, sequence-only route for injecting pairwise structural supervision into a protein language model via LoRA, with explicit attention to checkpoint selection and an honest statement of failure cases (local diagnostics, several ESM-S splits). The design is clean and the paper is candid about limitations. However, the headline claim—improvement on all eight protein-level tasks—currently rests on two unverified pillars: the statistical significance of small deltas in Table 2, and the absence of leakage between the 500k AlphaFold/Swiss-Prot adaptation set and downstream benchmarks derived from Swiss-Prot. The manuscript's own filtering audit (Figure 6) raises more questions than it answers because the final evaluation split is never identified. These issues are load-bearing and must be resolved before the central claim can be accepted.

major comments (3)
  1. [Supporting configuration and filtering analyses; Figure 6; Table 5] The manuscript never states whether Table 2 and Figures 3/8 used the original splits listed in Table 5 or the identity-filtered splits quantified in Figure 6. Figure 6 reports that exact-accession filtering retains only 9.0% of pooled evaluation entries and 30% identity filtering retains 1.7%, yet Table 5 lists full original partition sizes (e.g., 12,517 remote-homology test proteins). Because the adaptation set is AlphaFold/Swiss-Prot and several downstream benchmarks are also Swiss-Prot-derived, using unfiltered splits could explain the remote-homology gain as memorization of adaptation sequences rather than transferable contact supervision. The authors must state which split produced the final numbers and provide the corresponding results on the filtered split. This is the load-bearing gap for the all-eight-task claim.
  2. [Table 2] The claim 'LC-SEPLM improved all eight protein-level tasks' is not statistically supported for several tasks. Deltas of +0.0015 (EC), +0.0020 (subcellular localization), +0.0004 (membrane), and +0.0022 (HOU) are within or below the reported standard deviations (e.g., EC: 0.885186±0.005664 vs 0.883669±0.002586; HOU: 0.534068±0.016696 vs 0.531879±0.018151). Only remote homology, GO-MF, Pfam family, and strict membrane show deltas that clearly exceed the variability across seeds. The authors should either provide paired significance tests or appropriately weaken the claim to 'improved on the majority of protein-level tasks, with significant gains on...'.
  3. [Table 3; Discussion paragraph 'The comparison with ESM-S...'] The ESM-S comparison uses epoch 3 rather than the selected epoch-10 checkpoint, compares against official reported scores rather than a retrained control, and yields null/negative results on fold and GO-MF splits. The text acknowledges that differences 'may reflect both representation quality and benchmark design,' but the abstract still highlights a 'maximum absolute gain of 0.1771.' The abstract's phrasing is misleading because that gain is EC-only. The authors should either restrict the abstract claim to 'EC prediction' or provide a properly controlled comparison (same epoch, same training pipeline, retrained ESM-S).
minor comments (4)
  1. [Introduction, paragraph 4] The statement that a protein of length L contains O(L!) candidate residue pairs is incorrect; the number of pairs is O(L²). This typo should be fixed.
  2. [Figure 3 legend] The broken horizontal scale for remote homology is useful but should be explicitly described in the main text so readers do not infer a false visual magnitude for the other tasks.
  3. [Table 7] The exclusion of ClinVar and metal-ion diagnostics is reasonable, but the fact that archived ClinVar scores are exactly zero for every model suggests a possible data-path issue. The authors should audit these pipelines and state whether any other evaluation split or metric may be affected by a similar problem.
  4. [Data and code availability] The data availability statement says processed data are available 'upon reasonable request' and code will be released 'after publication.' For a paper whose central claim depends on filtering and checkpoint-selection details, the authors should provide reproducible code and processed filtered splits at submission, at least for review.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central comparison is an empirical benchmark evaluation, not a derivation that reduces to its inputs.

full rationale

The paper's central claim (all eight protein-level tasks improve over ESM2) rests on a controlled downstream evaluation: an ESM2 encoder is adapted with LoRA and a binary contact objective derived from AlphaFold/Swiss-Prot structures, and the frozen encoder is then evaluated on established protein-level benchmarks using mean-pooled sequence embeddings. The contact-loss definition (y_ij = 1(|i-j| >= 12 and C-alpha distance <= 8 Å)) is an auxiliary training target, not a function that computes the reported downstream metrics. No fitted parameter is renamed as a prediction; no equation in the paper defines the downstream macro-F1/micro-F1 values in terms of the adaptation objective. No self-citations by the present authors are load-bearing, and no uniqueness or ansatz result is imported from prior work by the same group. The manuscript itself flags the ESM-S comparison as a non-identical pipeline (Discussion: 'LC-SEPLM was evaluated against official ESM-S scores rather than retrained under an identical pipeline') and reports negative local diagnostics (ProteinGym, PTM-site), which weakens any claim of universal improvement but is not circularity. The only substantive concern raised in the text is the relationship between the 500k AlphaFold/Swiss-Prot adaptation set and the downstream benchmark splits: Figure 6 audits accession/identity filtering, while Table 5 lists the original split sizes and the paper does not state which filtered split generated Table 2. This is a potential data-leakage/external-validity problem, not a reduction of the result to the input by construction: the downstream labels are not used in adaptation, and the paper does not define the reported task scores as a deterministic function of the contact labels. Under the stated rules, such a benchmark-protocol ambiguity does not establish circularity. Score 0.

Assumptions & free parameters 8 free parameters · 4 assumptions · 0 invented entities

The central claim rests on hand-set contact definitions, LoRA hyperparameters, and the assumption that AlphaFold-derived labels plus identity filtering produce a clean and informative supervision signal. No new physical entities are introduced; the pair-query head is an architectural module removed at inference.

free parameters (8)
  • Long-range separation threshold = |i-j| >= 12
    Hand-set contact definition in Eq. 1; determines all positive/negative labels and therefore the supervision signal.
  • C-alpha distance threshold = <= 8 Å
    Hand-set contact definition; standard but arbitrary, and central to what counts as a positive contact.
  • LoRA rank = r = 4
    Chosen adaptation capacity; no sweep reported for this hyperparameter.
  • LoRA scaling factor = alpha = 8.0
    Part of the LoRA adaptation strength; chosen by design.
  • LoRA layer range = layers 22-29 (Tail-8)
    Selected from a one-epoch model-selection sweep (Fig. 7a); the early all-30-layer configuration was abandoned.
  • MLM loss weight = lambda = 0.6
    Selected from a two-epoch sweep (Fig. 7b); directly changes the balance between contact and sequence-modeling objectives.
  • Pair sampling limit = max 768 pairs/protein, ~2:1 neg:pos
    Controls label balance and computational cost; affects which contacts are seen during adaptation.
  • Sequence truncation length = 512 residues
    Truncation discards contacts beyond 512 residues and may bias supervision toward shorter proteins.
assumptions (4)
  • domain assumption AlphaFold-predicted Swiss-Prot structures are accurate enough that C-alpha-distance contacts are valid supervision labels.
    Used to construct all positive/negative contact labels (Eq. 1); the authors acknowledge predicted-structure bias as a limitation in the Discussion.
  • domain assumption The identity-filtering protocol separates adaptation training proteins from downstream benchmark evaluation proteins.
    Needed to prevent the reported gains from reflecting memorization of Swiss-Prot proteins; Figure 6 reports filtering, but the final protocol is ambiguous relative to Table 5.
  • domain assumption Mean-pooled frozen ESM2 residue embeddings provide adequate protein-level representations for downstream classifiers.
    All eight protein-level tasks use the mean-pooled 640-dimensional embedding described in Methods.
  • domain assumption The auxiliary MLM objective with lambda=0.6 preserves sequence modeling while contact supervision is learned.
    Validation MLM accuracy near 0.48 is cited as evidence of no collapse, but no comparison to the original ESM2 MLM accuracy is provided.

how reviews work

0 comments
Cite this review

Pith. "Pith review of LC-SEPLM: long-range contact-supervised adaptation for sequence-only protein representation learning." pith.science (2026). https://pith.science/paper/B7T257DQ

@misc{pith2026260722777,
  author       = {Pith},
  title        = {Pith review of: LC-SEPLM: long-range contact-supervised adaptation for sequence-only protein representation learning},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/B7T257DQ}},
  note         = {Machine review of arXiv:2607.22777}
}
read the original abstract

Protein language models learn transferable sequence representations. However, because they primarily model contextual dependencies along amino-acid sequences, their training objectives do not explicitly constrain the model to learn three-dimensional residue contacts formed after folding . Here, we introduce LC-SEPLM (Long-range Contact-supervised ESM Protein Language Model), which adapts ESM2 with LoRA and long-range residue-pair contact supervision while retaining sequence-only downstream inference. Pair-specific queries use cross-attention over the complete sequence to extract global sequence context associated with long-range spatial contacts. To expose the model to diverse structural information, we trained LC-SEPLM on 500,000 AlphaFold Swiss-Prot proteins. In downstream evaluation, LC-SEPLM improved all eight protein-level tasks relative to ESM2. The largest gain occurred in remote-homology recognition, where macro-F1 increased from 0.6122 to 0.6769 (+0.0647, or 6.47 percentage points). On the official ESM-S EC benchmark, LC-SEPLM also outperformed ESM-S with a maximum absolute gain of 0.1771. These results support residue-pair contact supervision as a bounded route for introducing structural information into protein sequence representations while preserving sequence-only inference.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

33 extracted references · 6 canonical work pages

  1. [1]

    Attention Is All You Need

    Vaswani A, Shazeer N, Parmar N, et al. Attention Is All You Need. Advances in Neural Information Processing Systems 30 (2017). https://arxiv.org/abs/1706.03762

    arXiv 2017

  2. [2]

    BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

    Devlin J, Chang MW, Lee K, Toutanova K. BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding. Proceedings of NAACL-HLT (2019). https://arxiv.org/abs/1810.04805

    arXiv 2019

  3. [3]

    Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences

    Rives A, Meier J, Sercu T, et al. Biological structure and function emerge from scaling unsupervised learning to 250 million protein sequences. Proceedings of the National Academy of Sciences 118, e2016239118 (2021). https://doi.org/10.1073/pnas.2016239118

    doi:10.1073/pnas.2016239118 2021

  4. [4]

    Evolutionary-scale prediction of atomic-level protein structure with a language model

    Lin Z, Akin H, Rao R, et al. Evolutionary-scale prediction of atomic-level protein structure with a language model. Science 379, 1123–1130 (2023). https://doi.org/10.1126/science.ade2574

    doi:10.1126/science.ade2574 2023

  5. [5]

    Unified rational protein engineering with sequence- based deep representation learning

    Alley EC, Khimulya G, Biswas S, et al. Unified rational protein engineering with sequence- based deep representation learning. Nature Methods 16, 1315–1322 (2019). https://doi.org/10.1038/s41592-019-0598-1

    doi:10.1038/s41592-019-0598-1 2019

  6. [6]

    Evaluating protein transfer learning with TAPE

    Rao R, Bhattacharya N, Thomas N, et al. Evaluating protein transfer learning with TAPE. Advances in Neural Information Processing Systems 32 (2019). https://proceedings.neurips.cc/paper/2019/hash/37f65c068b7723cd7809ee2d31d7861c- Abstract.html

    2019

  7. [7]

    ProtTrans: Toward Understanding the Language of Life Through Self-Supervised Learning

    Elnaggar A, Heinzinger M, Dallago C, et al. ProtTrans: Toward Understanding the Language of Life Through Self-Supervised Learning. IEEE Transactions on Pattern Analysis and Machine Intelligence 44, 7112–7127 (2022). https://doi.org/10.1109/TPAMI.2021.3095381

    arXiv 2022

  8. [8]

    Using deep learning to annotate the protein universe

    Bileschi ML, Belanger D, Bryant DH, et al. Using deep learning to annotate the protein universe. Nature Biotechnology 40, 932–937 (2022). https://doi.org/10.1038/s41587-021- 01179-w

    work page doi:10.1038/s41587-021- 2022

Show all 33 references
  1. [9]

    ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design

    Notin P, Kollasch A, Ritter D, et al. ProteinGym: Large-Scale Benchmarks for Protein Fitness Prediction and Design. Advances in Neural Information Processing Systems 36 (2023). https://proceedings.neurips.cc/paper_files/paper/2023/hash/cac723e5ff29f65e3fcbb0739ae91 bee-Abstrac...

    2023

  2. [10]

    FLIP: Benchmark tasks in fitness landscape inference for proteins

    Dallago C, Mou J, Johnston K, et al. FLIP: Benchmark tasks in fitness landscape inference for proteins. Advances in Neural Information Processing Systems 34, Datasets and Benchmarks Track (2021). https://datasets-benchmarks- proceedings.neurips.cc/paper/2021/hash/2b44928ae11fb...

    2021

  3. [11]

    Origins of coevolution between residues distant in protein 3D structures

    Anishchenko I, Ovchinnikov S, Kamisetty H, Baker D. Origins of coevolution between residues distant in protein 3D structures. Proceedings of the National Academy of Sciences 114, 9122–9127 (2017). https://doi.org/10.1073/pnas.1702664114

    2017 doi

  4. [12]

    Accurate de novo prediction of protein contact map by ultra-deep learning model

    Wang S, Sun S, Li Z, Zhang R, Xu J. Accurate de novo prediction of protein contact map by ultra-deep learning model. PLOS Computational Biology 13, e1005324 (2017). https://doi.org/10.1371/journal.pcbi.1005324

    2017 doi

  5. [13]

    Structure-Aware Transformer for Graph Representation Learning

    Chen D, O’Bray L, Borgwardt K. Structure-Aware Transformer for Graph Representation Learning. Proceedings of ICML / PMLR 162, 3469–3489 (2022). https://proceedings.mlr.press/v162/chen22r.html

    2022

  6. [14]

    Learning from protein structure with geometric vector perceptrons

    Jing B, Eismann S, Suriana PA, et al. Learning from protein structure with geometric vector perceptrons. International Conference on Learning Representations (2021). https://openreview.net/forum?id=1YLJDvSx6J4

    2021

  7. [15]

    Protein representation learning by geometric structure pretraining

    Zhang Z, Xu M, Jamasb AR, et al. Protein representation learning by geometric structure pretraining. International Conference on Learning Representations (2023). https://openreview.net/forum?id=to3qCB3tOh9

    2023

  8. [16]

    Endowing protein language models with structural knowledge

    Hartout P, Chen D, Pellizzoni P, Oliver C, Borgwardt K. Endowing protein language models with structural knowledge. Bioinformatics 41, btaf582 (2025). https://doi.org/10.1093/bioinformatics/btaf582

    2025 doi

  9. [17]

    Structure-based protein function prediction using graph convolutional networks

    Gligorijević V, Renfrew PD, Kosciolek T, et al. Structure-based protein function prediction using graph convolutional networks. Nature Communications 12, 3168 (2021). https://doi.org/10.1038/s41467-021-23303-9

    2021 doi

  10. [18]

    SaProt: Protein Language Modeling with Structure-aware Vocabulary

    Su J, Han C, Zhou Y, et al. SaProt: Protein Language Modeling with Structure-aware Vocabulary. International Conference on Learning Representations (2024). https://openreview.net/forum?id=6MRm3G4NiU

    2024

  11. [19]

    Fast and accurate protein structure search with Foldseek

    van Kempen M, Kim SS, Tumescheit C, et al. Fast and accurate protein structure search with Foldseek. Nature Biotechnology 42, 243–246 (2024). https://doi.org/10.1038/s41587-023- 01773-0

    2024 doi

  12. [20]

    ProstT5: Bilingual Language Model for Protein Sequence and Structure

    Heinzinger M, Weissenow K, Gomez Sanchez J, et al. ProstT5: Bilingual Language Model for Protein Sequence and Structure. bioRxiv (2023). https://doi.org/10.1101/2023.07.23.550085

    2023 doi

  13. [21]

    Structure-Informed Protein Language Model

    Zhang Z, Lu J, Chenthamarakshan V, et al. Structure-Informed Protein Language Model. arXiv:2402.05856 (2024). https://arxiv.org/abs/2402.05856

    2024 arXiv

  14. [22]

    Robust deep learning-based protein sequence design using ProteinMPNN

    Dauparas J, Anishchenko I, Bennett N, et al. Robust deep learning-based protein sequence design using ProteinMPNN. Science 378, 49–56 (2022). https://doi.org/10.1126/science.add2187

    2022 doi

  15. [23]

    Learning inverse folding from millions of predicted structures

    Hsu C, Verkuil R, Hie B, et al. Learning inverse folding from millions of predicted structures. Proceedings of the 39th International Conference on Machine Learning 162, 8946–8970 (2022). https://proceedings.mlr.press/v162/hsu22a.html

    2022

  16. [24]

    LoRA: Low-Rank Adaptation of Large Language Models

    Hu EJ, Shen Y, Wallis P, et al. LoRA: Low-Rank Adaptation of Large Language Models. International Conference on Learning Representations (2022). https://arxiv.org/abs/2106.09685

    2022 arXiv

  17. [25]

    Highly accurate protein structure prediction with AlphaFold

    Jumper J, Evans R, Pritzel A, et al. Highly accurate protein structure prediction with AlphaFold. Nature 596, 583–589 (2021). https://doi.org/10.1038/s41586-021-03819-2

    2021 doi

  18. [26]

    AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences

    Varadi M, Bertoni D, Magana P, et al. AlphaFold Protein Structure Database in 2024: providing structure coverage for over 214 million protein sequences. Nucleic Acids Research 52, D368–D375 (2024). https://doi.org/10.1093/nar/gkad1011

    2024 doi

  19. [27]

    UniProt: the Universal Protein Knowledgebase in 2023

    The UniProt Consortium. UniProt: the Universal Protein Knowledgebase in 2023. Nucleic Acids Research 51, D523–D531 (2023). https://doi.org/10.1093/nar/gkac1052

    2023 doi

  20. [28]

    ProteinShake: Building datasets and benchmarks for deep learning on protein structures

    Kucera T, Oliver C, Chen D, Borgwardt K. ProteinShake: Building datasets and benchmarks for deep learning on protein structures. Advances in Neural Information Processing Systems 36 (2023). https://proceedings.neurips.cc/paper_files/paper/2023/hash/b6167294ed3d6fc61e11e1592ce5...

    2023

  21. [29]

    Pfam: The protein families database in 2021

    Mistry J, Chuguransky S, Williams L, et al. Pfam: The protein families database in 2021. Nucleic Acids Research 49, D412–D419 (2021). https://doi.org/10.1093/nar/gkaa913

    2021 doi

  22. [30]

    The Gene Ontology resource: enriching a GOld mine

    Gene Ontology Consortium. The Gene Ontology resource: enriching a GOld mine. Nucleic Acids Research 49, D325–D334 (2021). https://doi.org/10.1093/nar/gkaa1113

    2021 doi

  23. [31]

    DeepLoc: prediction of protein subcellular localization using deep learning

    Almagro Armenteros JJ, Sønderby CK, Sønderby SK, et al. DeepLoc: prediction of protein subcellular localization using deep learning. Bioinformatics 33, 3387–3395 (2017). https://doi.org/10.1093/bioinformatics/btx431

    2017 doi

  24. [32]

    MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets

    Steinegger M, Söding J. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature Biotechnology 35, 1026–1028 (2017). https://doi.org/10.1038/nbt.3988 Tables Table 1 | Downstream benchmark composition. The table lists the task type, pre...

    2017 doi

  25. [33]

    LC-SEPLM scores are compared with reported ESM-S 150M results on EC, fold-classification and GO-MF splits

    Delta n EC level 1 macro-F1 0.883669 ± 0.002586 0.885186 ± 0.005664 +0.001517 5 Subcellular location macro-F1 0.759775 ± 0.005115 0.761797 ± 0.002369 +0.002022 5 Membrane binary macro-F1 0.917088 ± 0.004614 0.917534 ± 0.002996 +0.000446 5 Strict membrane binary macro-F1 0.9765...

Pith tools

Reviewed August 1, 2026 · model on record in the stance chip above.