Leveraging Multi-modal Representations to Predict Protein Melting Temperatures

Daiheng Zhang; Jinbo Xu; Xinyu Hong; Yan Zeng

REVIEW 5 major objections 5 minor 34 references

Leveraging Multi-modal Representations to Predict Protein Melting Temperatures

T0 review · 5 major / 5 minor · reviewed 2026-08-11 · deepseek-v4-flash

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

Pith's one-line read The paper claims that ESM3-DTm, a model built by fine-tuning the multimodal protein language model ESM3 on sequence and PDB-style structure inputs, predicts the change in protein melting temperature caused by a single-point mutation…

desk verdict The SOTA claim is confounded: ESM3-DTm sees ColabFold-predicted mutant structures while GeoDTm-3D does not, so the 0.03 PCC gap may be an input channel, not a better model. read the letter →

arxiv 2412.04526 v3 pith:WBRPPAV2 submitted 2024-12-05 cs.LG cs.CE

Daiheng Zhang , Yan Zeng , Xinyu Hong , Jinbo Xu This is my paper · ORCID

classification cs.LGcs.CE

keywords proteinmeltingtemperatureDeltaTmpredictionlanguagemodelESM3multimodalrepresentationstabilitymutationeffectfine-tuning

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

This paper claims that a protein language model that reads both amino-acid sequence and structure, fine-tuned end-to-end on a curated set of single-point mutations, predicts $\Delta T_m$ (the change in melting temperature caused by a mutation) better than any previously published method on the shared s571 benchmark. Their model, ESM3-DTm, built on the 1.4-billion-parameter ESM3, reports a Pearson correlation of 0.50, a mean absolute error of 5.21, and a root mean square error of 7.68 on 571 test mutations. The authors also run a fair comparison across ESM2, ESM3, SaProt, and OpenFold backbones to isolate which representation choices matter. The result matters because melting temperature is a direct experimental readout of protein stability, so better prediction would let protein engineers rank candidate mutations before going to the lab.

What carries the argument

The load-bearing object is ESM3-DTm, built on ESM3-1.4B, a multimodal protein language model that encodes each protein both as an amino-acid sequence and as a PDB-style structure. For a mutation, the model extracts a CLS summary embedding and a mutated-position token embedding for both the wild-type and mutant protein, then feeds them into two regression heads: one computes the outer product of the mutated-position embeddings, and the other concatenates layer-normalized differences of the CLS and position embeddings. The two heads' predictions are averaged, and the whole network is fine-tuned end-to-end with a mean-squared-error loss. The comparison variants (ESM2-DTm, SaProt-DTm, OpenFold-DTm) share the same two-head regression framework with their respective feature extractions.

What would settle it

Re-run the comparison with identical wild-type and mutant structures provided to ESM3-DTm and GeoDTm-3D, and compute bootstrap or repeated-seed confidence intervals on the Pearson correlation for the 571 test mutations; if those intervals overlap with 0.47, or if a sequence-identity check shows substantial overlap between the s4346 training set and s571 test set, the claimed improvement is not established.

Watch

Extended reading notes

Core claim

The paper's central claim is that jointly using sequence and structure through a multimodal, fully fine-tuned protein language model yields state-of-the-art $\Delta T_m$ prediction. On the s571 test set, ESM3-DTm surpasses the best prior structure-based method GeoDTm-3D (Pearson correlation 0.50 versus 0.47) and the best sequence-based method GeoDTm-Seq (0.46), while also beating its own sequence-only ESM3 variant (0.49) and an ESM2-based version (0.48). The authors attribute the gain to the multimodal ESM3 backbone and to fine-tuning the entire model rather than freezing the backbone, and they show that the choice of regression head substantially changes accuracy.

Load-bearing premise

The comparison rests on the assumption that the reported 0.50 Pearson correlation is genuinely better than the earlier predictor's 0.47 and is not a fluke of evaluation noise, of predicted mutant structures the earlier method never saw, or of training proteins too similar to test proteins; the paper gives no confidence intervals and no sequence-overlap check.

Editorial extensions

If this is right

On the shared s571 benchmark, ESM3-DTm reports the best published $\Delta T_m$ prediction: Pearson correlation 0.50, MAE 5.21, and RMSE 7.68.
Multimodal input helps: ESM3-DTm (0.50) outperforms the same model with sequence-only input (0.49) and the ESM2-based model (0.48).
Fully fine-tuning the backbone is better than freezing it: ESM2-DTm improves from Pearson 0.46 to 0.48 when the backbone is trained.
Regression-head design matters: combining mutated-position outer product with CLS and position differences (0.41/0.40) far outperforms simple token concatenation (0.21) on the ESM2 backbone.

Reading between the lines

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

If the small sequence-versus-multimodal gap (0.49 versus 0.50) survives fresh evaluation, the practical lesson is that most of the predictive signal lives in the language model and the regression setup, with structure adding only a modest increment on this benchmark.
The same two-head ensemble could be transferred to other mutation-effect regression tasks such as $\Delta\Delta G$ or enzyme kinetic parameters, where multimodal ESM3 embeddings may behave differently than on melting temperature.
A decisive robustness test would be to split the available mutations by sequence clusters at 30% identity and retrain from scratch, reporting confidence intervals; that would show whether the reported ordering of backbones is stable or tied to the particular s4346/s571 partition.

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.

Desk Editor's Note

read the letter

The paper applies ESM3 to ΔTm prediction and reports a new SOTA (PCC 0.50 vs 0.47 for GeoDTm-3D). The stress-test note is right: that comparison is not apples-to-apples. ESM3-DTm takes both sequence and PDB structure inputs, and the authors generated ColabFold structures for all mutated proteins. GeoDTm-3D, per the cited work, uses only wild-type structures. So ESM3-DTm gets a predicted structure of the very mutant whose ΔTm it is predicting, an input that already encodes the mutation and any stability-relevant structural shift. A 0.03 PCC gain could easily come from that extra channel, and with no confidence intervals we cannot tell. The abstract's phrase "fair evaluation" does not survive contact with this setup.

That said, the paper does real work. The systematic comparison of ESM2, ESM3, SaProt, and OpenFold backbones under similar training protocols is useful, and the authors were careful to build a training/validation split with MMseqs2 at 50% sequence identity, which is more than some prior work bothered with. The ablations of regression heads and fine-tuning vs frozen backbones are informative. The external s571 test set is independent, and the reported metrics support the ranking among the models they trained themselves.

The soft spots beyond the mutant-structure asymmetry are several but manageable. No error bars or significance tests anywhere. The OpenFold comparison freezes the backbone while the PLMs are fully fine-tuned, which is not a fair representation comparison. Algorithm 1 is cryptic: the outer-product head is O(d^3) in parameters, which for ESM3's hidden dimension is enormous, and the paper never discusses whether that head is the reason ESM3 wins or just a memory hazard. No code or data released, so none of this is independently reproducible. They also never check sequence identity between s571 and their training set, so leakage cannot be ruled out, though the external origin makes it less likely.

My bottom line: the paper is an honest empirical exploration, but the headline SOTA claim is not established. The right fix is an ablation where ESM3-DTm is given only wild-type structure (or none), and ideally confidence intervals from multiple seeds. If the margin over GeoDTm-3D vanishes, the paper is still a reasonable benchmark comparison, just not a SOTA claim.

I would not desk-reject it. The task is niche but practical, the experiments are mostly sensible, and the issues are addressable in revision. I would send it to review with a specific request for the wild-type-only control and error bars.

Referee Report

5 major / 5 minor

Summary. The paper introduces ESM3-DTm, a regression model for predicting changes in protein melting temperature (ΔTm) upon single-point mutations. The model fine-tunes the ESM3-1.4B multimodal language model on both wild-type and mutated protein sequences and, in its full form, on predicted PDB structures of the mutant, using a regression head that combines an outer-product of mutated-position embeddings with a linear combination of CLS embeddings. Training uses the s4346 set and evaluation uses the s571 test set from the GeoStab benchmark. The authors report a Pearson correlation of 0.50, MAE of 5.21, and RMSE of 7.68, which they claim is state-of-the-art compared with GeoDTm-3D (PCC 0.47). The paper also compares ESM2, SaProt, and OpenFold backbones under nominally similar settings and includes ablations of regression heads and fine-tuning strategies.

Significance. If the reported comparison is sound, the paper would provide a useful demonstration that multimodal protein language model representations can improve ΔTm prediction, a task with relatively few deep-learning baselines. The use of an external test set (s571), the systematic comparison of several backbone models, and the ablation studies are strengths that make the central claim easy to scrutinize. However, the headline SOTA claim is currently supported only by a small margin over a baseline that does not receive the same mutant-structure information, with no uncertainty quantification. The paper is therefore of moderate significance; the result is plausible but not yet established.

major comments (5)

[Data; Algorithm 1; Table 1] The SOTA comparison is confounded by an information asymmetry. The Data section states that 'for all mutated structures, we generated PDB structures using ColabFold,' and ESM3-DTm is described as accepting both sequence and PDB structure inputs. GeoDTm-3D, the baseline to which the headline PCC improvement is compared, is a structure-based method that, per the cited work (Xu et al. 2023), uses wild-type structures and is not given the mutant structure. Thus ESM3-DTm has access to a predicted structure of the very mutant whose ΔTm is being predicted, which is an additional input channel that may encode stability-relevant information independent of the model architecture. The reported margin (PCC 0.50 vs. 0.47) could plausibly be explained by this extra channel. Please provide a controlled comparison: either supply the same mutant structures to GeoDTm-3D (or an equivalent structure-based baseline), or restrict ESM3-DTm to wild-type structures only, and report whether the PCC gap persists.
[Results; Tables 1 and 2] No uncertainty quantification is provided for the headline metrics. The improvements over GeoDTm-3D are small in absolute terms (0.03 PCC, 0.10 MAE, 0.35 RMSE), and the 571 test mutations are distributed across only 37 proteins, so the effective sample size is considerably smaller than 571. Without confidence intervals, bootstrap estimates, or a paired significance test, the claim that ESM3-DTm is statistically superior to GeoDTm-3D is not supported. Please report per-protein or bootstrapped intervals and a paired comparison of the PCC difference.
[Experiments: Model Setup; Table 2] The 'fair evaluation' between backbones is not actually controlled. OpenFold-DTm freezes the backbone and trains only the linear layer, while ESM2, ESM3, and SaProt backbones are fully fine-tuned; ESM3-DTm additionally receives mutant structures, while ESM2-DTm and ESM3-DTm (seq only) do not. The performance differences in Table 2 therefore conflate architecture, input modality, and fine-tuning strategy. Please equalize the training regimes (e.g., fine-tune all backbones or freeze all backbones) and control the structure input before drawing conclusions about the relative merits of the backbone models.
[Algorithm 1] The outer-product regression head is written with a linear layer of O(d^3) parameters. In Algorithm 1, Head1 = Flatten(am ⊗ aw ∈ R^{d^2}) with a learned map W: R^{d^2} → R^d gives a weight matrix of dimension d^2 × d = d^3. For a multimodal hidden dimension on the order of 1536, this layer would contain roughly 3.6 billion parameters, exceeding the size of the ESM3-1.4B backbone. Please clarify whether this is the actual implementation, and if so, describe the regularization used to control overfitting; alternatively, correct the notation if a lower-dimensional projection or low-rank approximation is employed.
[Data; data split] There is no sequence-identity control between the training and test sets. The authors use MMseqs2 at 50% identity to create the validation split, but they do not report whether proteins in s571 share high sequence identity with proteins in s4346. If such overlap exists, the reported metrics would be optimistic relative to a truly held-out setting. Please report the maximum pairwise sequence identity between each s571 protein and the s4346 training set, and consider presenting results on a subset clustered at, for example, 30% identity.

minor comments (5)

[Abstract] The abstract mentions AlphaFold, but the experiments use OpenFold as the structure-prediction backbone; please align the terminology.
[Algorithm 1] The definitions of N1 and N2 are incomplete: the algorithm states that N1 and N2 are linear layers connected after Head1 and Head2, but their output dimensions are not specified.
[Table 4] Table 4 lacks the upward/downward arrows used in other tables to indicate whether higher or lower values are better; please make the table headers consistent.
[Regression Head] The cross-reference 'in Section' in the Model Setup paragraph is unresolved; please insert the correct section number.
[Reproducibility] No code or data availability statement is provided; given the central role of ColabFold-generated structures and ESM3 inference, a reproducibility statement would substantially strengthen the paper.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity; the ΔTm predictions are trained on s4346 and evaluated on the external s571 benchmark with no fitted parameter derived from test labels.

full rationale

The paper's derivation chain is a standard supervised regression setup. The model is trained on the s4346 set and evaluated on the s571 set; hyperparameters are chosen using an MMseqs2-based train/validation split and the final model is retrained on combined training and validation data, none of which uses test labels. The ESM3-DTm regression heads (outer product and linear combination of position/CLS embeddings, Algorithm 1) are learned functions of the input embeddings, not definitions of the target ΔTm. The SOTA comparison to GeoDTm-3D is empirical, with all baseline numbers quoted from an external source, and the performance gap is not forced by construction within this paper. The only notable concern is an evaluation-fairness asymmetry, not circularity: the paper states that 'for all mutated structures, we generated PDB structures using ColabFold' and that ESM3 accepts 'both sequence and PDB structure inputs,' whereas the GeoDTm-3D baseline may not receive a mutant structure; this could confound the comparison, but it does not make the prediction equivalent to its input or reduce the result to a fitted parameter renamed as a prediction. Consequently, no circular step is identifiable under the specified criteria.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claim rests on the adequacy of ESM3 representations, the quality of predicted structures, and the integrity of the external benchmark. These are domain assumptions rather than proven facts. No new entities are postulated.

free parameters (5)

learning_rate = 1e-5
Set manually; affects convergence and final performance.
num_epochs = 10
Chosen for all models; no early stopping described.
gradient_clip_norm = 0.1
Set manually for stable training.
mmseqs2_sequence_identity = 50%
Used to split training/validation; threshold chosen by hand and affects hyperparameter selection.
train_val_split_ratio = 8:2
Split ratio chosen by hand.

assumptions (4)

domain assumption ESM3 embeddings capture sequence and structure features relevant to thermal stability changes.
The entire framework rests on the premise that PLM representations are informative for DeltaTm.
domain assumption ColabFold-predicted structures are sufficiently accurate substitutes for experimental PDB structures for both wild-type and mutant proteins.
The paper uses ColabFold-generated PDBs for proteins without PDB IDs and for all mutants; if these predictions are poor, the multimodal input may be misleading.
domain assumption The s571 test set labels are accurate and the test set is not contaminated by training-set homologs.
Performance claims depend on the integrity of the external benchmark; the paper does not re-verify the split.
standard math Standard transformer and optimization mathematics (backpropagation, Adam) are correct.
Relying on standard ML machinery implicitly.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Leveraging Multi-modal Representations to Predict Protein Melting Temperatures." pith.science (2026). https://pith.science/paper/WBRPPAV2

@misc{pith2026241204526,
  author       = {Pith},
  title        = {Pith review of: Leveraging Multi-modal Representations to Predict Protein Melting Temperatures},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/WBRPPAV2}},
  note         = {Machine review of arXiv:2412.04526}
}

read the original abstract

Accurately predicting protein melting temperature changes (Delta Tm) is fundamental for assessing protein stability and guiding protein engineering. Leveraging multi-modal protein representations has shown great promise in capturing the complex relationships among protein sequences, structures, and functions. In this study, we develop models based on powerful protein language models, including ESM-2, ESM-3 and AlphaFold, using various feature extraction methods to enhance prediction accuracy. By utilizing the ESM-3 model, we achieve a new state-of-the-art performance on the s571 test dataset, obtaining a Pearson correlation coefficient (PCC) of 0.50. Furthermore, we conduct a fair evaluation to compare the performance of different protein language models in the Delta Tm prediction task. Our results demonstrate that integrating multi-modal protein representations could advance the prediction of protein melting temperatures.

Figures

Figures reproduced from arXiv: 2412.04526 by the authors.

**Figure 1.** Model Architecture. ESM3-DTm efficiently predicts ∆Tm. We also present ESM2-DTm, Saprot-DTm, and OpenfoldDTm here. “I4A” means mutation from I to A at position 4. ture extractor we employ. We extract sequence features Pw and Pm from the Evoformer and Structure Module, and also the embeddings at the specific mutated positions, aw and am. These features are then pass through a linear layer. We train the model using t… view at source ↗

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

34 extracted references · 27 canonical work pages

[1]

J.; Berenberg, D.; Fisk, I.; Zanichelli, N.; et al

Ahdritz, G.; Bouatta, N.; Floristean, C.; Kadyan, S.; Xia, Q.; Gerecke, W.; O’Donnell, T. J.; Berenberg, D.; Fisk, I.; Zanichelli, N.; et al. 2024. OpenFold: Retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization. Nature Methods, 1--11

work page 2024
[2]

Benevenuta, S.; Pancotti, C.; Fariselli, P.; Birolo, G.; and Sanavia, T. 2021. An antisymmetric neural network to predict free energy changes in protein variants. Journal of Physics D: Applied Physics, 54(24): 245403

work page 2021
[3]

Bepler, T.; and Berger, B. 2021. Learning the protein language: Evolution, structure, and function. Cell systems, 12(6): 654--669

work page 2021
[4]

Chen, Y.; Xu, Y.; Liu, D.; Xing, Y.; and Gong, H. 2024. An end-to-end framework for the prediction of protein structure and fitness from single sequence. Nature Communications, 15(1): 7400

work page 2024
[5]

Diederik, P. K. 2014. Adam: A method for stochastic optimization. (No Title)

2014
[6]

M.; and Fields, S

Fowler, D. M.; and Fields, S. 2014. Deep mutational scanning: a new style of protein science. Nature methods, 11(8): 801--807

work page 2014
[7]

M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; Prabakaran, P.; and Sarai, A

Gromiha, M. M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; Prabakaran, P.; and Sarai, A. 2000. ProTherm, version 2.0: thermodynamic database for proteins and mutants. Nucleic acids research, 28(1): 283--285

work page 2000
[8]

M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; and Sarai, A

Gromiha, M. M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; and Sarai, A. 1999. ProTherm: thermodynamic database for proteins and mutants. Nucleic acids research, 27(1): 286--288

work page 1999

Show all 34 references

[9]

M.; Uedaira, H.; An, J.; Selvaraj, S.; Prabakaran, P.; and Sarai, A

Gromiha, M. M.; Uedaira, H.; An, J.; Selvaraj, S.; Prabakaran, P.; and Sarai, A. 2002. ProTherm, thermodynamic database for proteins and mutants: developments in version 3.0. Nucleic acids research, 30(1): 301--302

2002
[10]

J.; Oktay, D.; Lin, Z.; Verkuil, R.; Tran, V

Hayes, T.; Rao, R.; Akin, H.; Sofroniew, N. J.; Oktay, D.; Lin, Z.; Verkuil, R.; Tran, V. Q.; Deaton, J.; Wiggert, M.; et al. 2024. Simulating 500 million years of evolution with a language model. bioRxiv, 2024--07

2024
[11]

L.; Shanker, V

Hie, B. L.; Shanker, V. R.; Xu, D.; Bruun, T. U.; Weidenbacher, P. A.; Tang, S.; Wu, W.; Pak, J. E.; and Kim, P. S. 2024. Efficient evolution of human antibodies from general protein language models. Nature Biotechnology, 42(2): 275--283

2024
[12]

Jumper, J.; Evans, R.; Pritzel, A.; Green, T.; Figurnov, M.; Ronneberger, O.; Tunyasuvunakool, K.; Bates, R.; Z \' dek, A.; Potapenko, A.; et al. 2021. Highly accurate protein structure prediction with AlphaFold. nature, 596(7873): 583--589

2021
[13]

S.; Bava, K

Kumar, M. S.; Bava, K. A.; Gromiha, M. M.; Prabakaran, P.; Kitajima, K.; Uedaira, H.; and Sarai, A. 2006. ProTherm and ProNIT: thermodynamic databases for proteins and protein--nucleic acid interactions. Nucleic acids research, 34(suppl\_1): D204--D206

2006
[14]

K.; Kerkhoven, E

Li, F.; Yuan, L.; Lu, H.; Li, G.; Chen, Y.; Engqvist, M. K.; Kerkhoven, E. J.; and Nielsen, J. 2022. Deep learning-based k cat prediction enables improved enzyme-constrained model reconstruction. Nature Catalysis, 5(8): 662--672

2022
[15]

Lin, Z.; Akin, H.; Rao, R.; Hie, B.; Zhu, Z.; Lu, W.; dos Santos Costa, A.; Fazel-Zarandi, M.; Sercu, T.; Candido, S.; et al. 2022. Language models of protein sequences at the scale of evolution enable accurate structure prediction. BioRxiv, 2022: 500902

2022
[16]

Masso, M.; and Vaisman, I. 2014. AUTO-MUTE 2.0: a portable framework with enhanced capabilities for predicting protein functional consequences upon mutation. Adv Bioinf. 2014

2014
[17]

Masso, M.; and Vaisman, I. I. 2008. Accurate prediction of stability changes in protein mutants by combining machine learning with structure based computational mutagenesis. Bioinformatics, 24(18): 2002--2009

2008
[18]

Mirdita, M.; Sch \"u tze, K.; Moriwaki, Y.; Heo, L.; Ovchinnikov, S.; and Steinegger, M. 2022. ColabFold: making protein folding accessible to all. Nature methods, 19(6): 679--682

2022
[19]

a henb \

Ouyang-Zhang, J.; Diaz, D.; Klivans, A.; and Kr \"a henb \"u hl, P. 2024. Predicting a protein's stability under a million mutations. Advances in Neural Information Processing Systems, 36

2024
[20]

P.; and Blundell, T

Pandurangan, A. P.; and Blundell, T. L. 2020. Prediction of impacts of mutations on protein structure and interactions: SDM, a statistical approach, and mCSM, using machine learning. Protein Science, 29(1): 247--257

2020
[21]

E.; Ascher, D

Pires, D. E.; Ascher, D. B.; and Blundell, T. L. 2014. mCSM: predicting the effects of mutations in proteins using graph-based signatures. Bioinformatics, 30(3): 335--342

2014
[22]

Pucci, F.; Bourgeas, R.; and Rooman, M. 2016. Predicting protein thermal stability changes upon point mutations using statistical potentials: Introducing HoTMuSiC. Scientific reports, 6(1): 23257

2016
[23]

Steinegger, M.; and S \"o ding, J. 2017. MMseqs2 enables sensitive protein sequence searching for the analysis of massive data sets. Nature biotechnology, 35(11): 1026--1028

2017
[24]

Su, J.; Han, C.; Zhou, Y.; Shan, J.; Zhou, X.; and Yuan, F. 2023. Saprot: Protein language modeling with structure-aware vocabulary. bioRxiv, 2023--10

2023
[25]

J.; Mangan, N

Tsuboyama, K.; Dauparas, J.; Chen, J.; Laine, E.; Mohseni Behbahani, Y.; Weinstein, J. J.; Mangan, N. M.; Ovchinnikov, S.; and Rocklin, G. J. 2023. Mega-scale experimental analysis of protein folding stability in biology and design. Nature, 620(7973): 434--444

2023
[26]

I.; Strashnov, P

Umerenkov, D.; Shashkova, T. I.; Strashnov, P. V.; Nikolaev, F.; Sindeeva, M.; Ivanisenko, N. V.; and Kardymon, O. L. 2022. PROSTATA: protein stability assessment using transformers. BioRxiv, 2022--12

2022
[27]

S.; Tumescheit, C.; Mirdita, M.; Gilchrist, C

van Kempen, M.; Kim, S. S.; Tumescheit, C.; Mirdita, M.; Gilchrist, C. L.; S \"o ding, J.; and Steinegger, M. 2022. Foldseek: fast and accurate protein structure search. Biorxiv, 2022--02

2022
[28]

C.; and Lesk, A

Whisstock, J. C.; and Lesk, A. M. 2003. Prediction of protein function from protein sequence and structure. Quarterly reviews of biophysics, 36(3): 307--340

2003
[29]

S.; Nguyen, T.-B.; Karmarkar, M.; Portelli, S.; Rezende, P

Xavier, J. S.; Nguyen, T.-B.; Karmarkar, M.; Portelli, S.; Rezende, P. M.; Velloso, J. P.; Ascher, D. B.; and Pires, D. E. 2021. ThermoMutDB: a thermodynamic database for missense mutations. Nucleic acids research, 49(D1): D475--D479

2021
[30]

Xu, Y.; Liu, D.; and Gong, H. 2023. Improving the prediction of protein stability changes upon mutations by geometric learning and a pre-training strategy. bioRxiv, 2023--05

2023
[31]

D.; and Luo, X

Yu, H.; Deng, H.; He, J.; Keasling, J. D.; and Luo, X. 2023. UniKP: a unified framework for the prediction of enzyme kinetic parameters. Nature communications, 14(1): 8211

2023
[32]

Zhang, Q.; Ding, K.; Lyv, T.; Wang, X.; Yin, Q.; Zhang, Y.; Yu, J.; Wang, Y.; Li, X.; Xiang, Z.; et al. 2024. Scientific large language models: A survey on biological & chemical domains. arXiv preprint arXiv:2401.14656

2024 arXiv
[33]

, " * write output.state after.block = add.period write newline

ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...
[34]

write newline

" write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.

[1] [1]

J.; Berenberg, D.; Fisk, I.; Zanichelli, N.; et al

Ahdritz, G.; Bouatta, N.; Floristean, C.; Kadyan, S.; Xia, Q.; Gerecke, W.; O’Donnell, T. J.; Berenberg, D.; Fisk, I.; Zanichelli, N.; et al. 2024. OpenFold: Retraining AlphaFold2 yields new insights into its learning mechanisms and capacity for generalization. Nature Methods, 1--11

work page 2024

[2] [2]

Benevenuta, S.; Pancotti, C.; Fariselli, P.; Birolo, G.; and Sanavia, T. 2021. An antisymmetric neural network to predict free energy changes in protein variants. Journal of Physics D: Applied Physics, 54(24): 245403

work page 2021

[3] [3]

Bepler, T.; and Berger, B. 2021. Learning the protein language: Evolution, structure, and function. Cell systems, 12(6): 654--669

work page 2021

[4] [4]

Chen, Y.; Xu, Y.; Liu, D.; Xing, Y.; and Gong, H. 2024. An end-to-end framework for the prediction of protein structure and fitness from single sequence. Nature Communications, 15(1): 7400

work page 2024

[5] [5]

Diederik, P. K. 2014. Adam: A method for stochastic optimization. (No Title)

2014

[6] [6]

M.; and Fields, S

Fowler, D. M.; and Fields, S. 2014. Deep mutational scanning: a new style of protein science. Nature methods, 11(8): 801--807

work page 2014

[7] [7]

M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; Prabakaran, P.; and Sarai, A

Gromiha, M. M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; Prabakaran, P.; and Sarai, A. 2000. ProTherm, version 2.0: thermodynamic database for proteins and mutants. Nucleic acids research, 28(1): 283--285

work page 2000

[8] [8]

M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; and Sarai, A

Gromiha, M. M.; An, J.; Kono, H.; Oobatake, M.; Uedaira, H.; and Sarai, A. 1999. ProTherm: thermodynamic database for proteins and mutants. Nucleic acids research, 27(1): 286--288

work page 1999