bio.rodeo
ModelsOrganizationsProvidersLeaderboardAboutSign in
bio.rodeo

The authoritative source for evaluating biological foundation models. No hype, just honest analysis.

Categories
  • DNA & Gene
  • RNA
  • Protein
  • Small molecule
  • Single-cell
  • Spatial omics
  • Pathology
  • Imaging
  • Metabolomics
  • Biosignals
  • Language model
bio.rodeoModelsOrganizationsProvidersLeaderboardAboutFAQSubmit a modelTermsPrivacyContact
© 2026 Pulsatance. All rights reserved. ~
Built by Pulsatance
models / protein / esmstabp
Protein
Iowa State UniversityReleased February 2025

ESMStabP

Protein melting temperature prediction from sequence, pairing frozen ESM2 embeddings with growth-temperature and thermophilicity features.

The short version

  • —Estimates melting temperature from sequence alone, skipping calorimetry and CD spectroscopy
  • —Layer 33 was chosen after testing all 33 ESM2 layers in isolation
  • —Adds growth temperature and thermophilic class on top of the language model signal
  • —Screens enzyme variants for stability before committing sequences to the bench
34Openness

Where to run it

No providers recorded yet. Browse all providers

ESMStabP predicts a protein's melting temperature (Tm) directly from its amino acid sequence. It was developed by Marcus Ramos, Robert L. Jernigan, and Mesih Kilinc at Iowa State University and posted to bioRxiv in February 2025.

Thermostability determines whether a protein survives the temperatures encountered during cultivation, processing, and storage, which makes it a gating property for industrial enzymes, biologics formulation, and food science applications. Measuring it experimentally is slow and costly: differential scanning calorimetry and circular dichroism spectroscopy are labor-intensive per protein, and the resulting scarcity of labeled data has been the field's binding constraint as much as any modeling limitation.

ESMStabP builds directly on DeepSTABp, an earlier regression model for the same task, and makes two changes. First, it replaces the sequence representation with embeddings from a frozen ESM-2 protein language model, and identifies empirically which layer of that model carries the relevant signal. Second, it enlarges the training data by consolidating and cleaning the datasets used by several prior thermostability models into a single deduplicated corpus, which addresses the data scarcity problem rather than working around it.

#Key Features

  • Empirical layer selection: All 33 layers of the ESM2 encoder were tested in isolation as regression inputs; layer 33 performed best by a clear margin, and only those embeddings are used. The authors argue this layer-isolation procedure should generalize to other ESM2-based predictors.
  • Consolidated training corpus: Datasets from multiple published thermostability models were combined, deduplicated, and harmonized into one set, then balanced by downsampling the over-represented non-thermophilic proteins.
  • Auxiliary biological features: Optimal growth temperature, experimental condition (whole cell versus lysate), and a thermophilic/non-thermophilic classification are supplied alongside the embedding — each was ablated individually, and using all of them together outperformed any subset.
  • Frozen encoder, light head: The language model is not fine-tuned; a random forest regressor sits on top, chosen after comparison against linear regression, polynomial regression, and support vector regression.
  • Hosted inference: A deployed prediction endpoint is available through BioLM, so a sequence can be scored without local installation.

#Technical Details

Sequences are encoded with the esm2_t33_650M_UR50 checkpoint and truncated to a maximum of 1,022 residues to fit the model's context. Layer-33 output logits are pooled into a fixed-dimension embedding, concatenated with the three auxiliary features, and passed to a random forest regressor. Labels come from high-throughput mass-spectrometry thermal proteome profiling assays; the thermophilic classification follows the convention used by TemBERTure, with growth temperature above 60°C marking thermophilic and below 30°C non-thermophilic. The corpus was split 80/20 for training and testing, with 5-fold cross-validation.

In a head-to-head comparison where all models were retrained on the identical balanced dataset, ESMStabP reached R² 0.94 and Pearson correlation 0.92, against DeepSTABp at 0.81 and 0.88 and ProTstab2 at 0.51 and 0.68. On those balanced-set error metrics ESMStabP's MAE of 3.42 was better than DeepSTABp's 3.62, though its MSE and RMSE were slightly worse. Trained on the original unbalanced corpus it improved across all five metrics, reaching MSE 13.71, RMSE 3.70, and MAE 2.79. The paper's headline figures are R² 0.95 and PCC 0.97. Ablations showed that dropping either optimal growth temperature or the thermophilic class reduced PCC to 0.87 and 0.90 respectively, so both contribute despite being closely related quantities.

#Applications

The practical use is triage in protein engineering: ranking candidate enzyme variants by predicted stability so that only the promising ones consume bench time, which matters most in directed evolution campaigns and industrial biocatalyst development where variant libraries are large. Proteome-scale stability surveys are another fit, since scoring is cheap enough to run across an organism's full complement of proteins. Formulation and process scientists can use predicted Tm to anticipate which proteins are at risk during heating steps. The hosted endpoint means groups without local GPU capacity can still obtain predictions.

#Impact

ESMStabP's most transferable finding is not the model but the layer-isolation result: performance varied considerably across ESM2 layers, and the common practice of taking the final layer by default may leave signal unused in other property predictors built the same way. The architecture itself is deliberately modest — a random forest on frozen embeddings — which keeps it cheap to train and easy to interpret relative to end-to-end deep models. Several caveats apply. The work remains a preprint, not peer-reviewed. The code is public but unlicensed: the authors' repository holds the dataset-assembly pipeline, its three base corpora, the training script, and LoRA fine-tuning experiments, and a companion Flask repository ships the four fitted regressors — but neither carries a licence, so the artifacts are inspectable rather than legally reusable. The reported R² and PCC figures differ between the abstract and the comparison table, and the auxiliary features it depends on — optimal growth temperature in particular — are not always available for an arbitrary sequence of interest. Finally, training labels derive from thermal proteome profiling, whose melting temperatures reflect cellular context and do not always match purified-protein measurements.

At a glance

Released
February 2025
Category
Protein
Organization
Iowa State University

Links

GitHub RepositoryResearch PaperDemo

Tags

embeddingsenzymesprotein_engineeringprotein_stability_predictionproteomicsrandom_foresttransfer_learningtransformer

Something wrong?

Much of this page is generated or calculated automatically. Flag anything that looks off and we will re-run it.