TUnA: an uncertainty-aware transformer model for sequence-based protein–protein interaction prediction
Open accessThe DOI matches its Crossref record; 10 of the 10 references it cites were found in a registry or an index.
Abstract
Protein–protein interactions (PPIs) are important for many biological processes, but predicting them from sequence data remains challenging. Existing deep learning models often cannot generalize to proteins not present in the training set and do not provide uncertainty estimates for their predictions. To address these limitations, we present TUnA, a Transformer-based uncertainty-aware model for PPI prediction. TUnA uses ESM-2 embeddings with Transformer encoders and incorporates a Spectral-normalized Neural Gaussian Process. TUnA achieves state-of-the-art performance and, importantly, evaluates uncertainty for unseen sequences. We demonstrate that TUnA’s uncertainty estimates can effectively identify the most reliable predictions, significantly reducing false positives. This capability is crucial in bridging the gap between computational predictions and experimental validation.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Szklarczyk, D., Gable, A. L., Nastou, K. C., Lyon, D., Kirsch, R., Pyysalo, S., Doncheva, N. T., Legeay, M., Fang, T., Bork, P., et al. (2021) · Nucleic Acids Research · doi:10.1093/nar/gkaa1074 · Cited by 8,056 · cited 1×
“The cross-species dataset (Supplementary Table 4) was constructed by Sledzieski et al ., with PPI data originating from the STRING database (v11) filtered to only include experimentally determined physical binding interactions [7, 27].”
MethodDropout as a bayesian approximation: Representing model uncertainty in deep learning
Gal, Y., & Ghahramani, Z. (2015) · arXiv · doi:10.48550/arxiv.1506.02142 · Cited by 6,471 · cited 1×
“Compared to other uncertainty estimation methods such as Deep Ensembles [25] and Monte Carlo Dropout [26], SNGP only requires only a single network, thus offering a low-cost uncertainty estimate, and also combines the flexibility of neural network models and better uncertainty calibration of GP.”
MethodEvolutionary-scale prediction of atomic level protein structure with a language model
Lin, Z., Akin, H., Rao, R., Hie, B., Zhu, Z., Lu, W., Smetanin, N., Verkuil, R., Kabeli, O., Shmueli, Y., et al. (2022) · doi:10.1101/2022.07.20.500902 · Cited by 5,990
On calibration of modern neural networks
Guo, C., Pleiss, G., Sun, Y., & Weinberger, K. Q. (2017) · arXiv · doi:10.48550/arxiv.1706.04599 · Cited by 3,435
Deep Neural Networks are Easily Fooled: High Confidence Predictions for Unrecognizable Images
Nguyen, A., Yosinski, J., & Clune, J. (2014) · arXiv · doi:10.48550/arxiv.1412.1897 · Cited by 3,258
Deep neural networks are easily fooled: High confidence predictions for unrecognizable images
Nguyen, A., Yosinski, J., & Clune, J. (2015) · 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2015.7298640 · Cited by 3,258
Searching for activation functions
Ramachandran, P., Zoph, B., & Le, Q. V. (2017) · arXiv · doi:10.48550/arxiv.1710.05941 · Cited by 2,335 · cited 1×
“Additionally, we replace all instances of the ReLU activation function with the Swish activation function, as Ramachandran et al. have shown its effectiveness over ReLU [14].”
MethodChen, L., Tan, X., Wang, D., Zhong, F., Liu, X., Yang, T., Luo, X., Chen, K., Jiang, H., & Zheng, M. (2020) · Bioinformatics · doi:10.1093/bioinformatics/btaa524 · Cited by 607
Learning the protein language: Evolution, structure, and function
Bepler, T., & Berger, B. (2021) · Cell Systems · doi:10.1016/j.cels.2021.05.017 · Cited by 483 · cited 2×
“In addition, while Topsy-Turvy utilizes the ProSE [29] protein language model’s N × 6165 embedding for each length N protein sequence, TUnA uses ESM-2’s N × 640 embedding, requiring ∼ 10 times less memory.”
BackgroundMethod
Cited by
Predicting protein-protein interactions in the human proteome
Zhang, J., Humphreys, I. R., Pei, J., Kim, J., Choi, C., Yuan, R., Durham, J., Liu, S., Choi, H.-J., Baek, M., et al. (2025) · Science · doi:10.1126/science.adt1630 · Cited by 87
A paired sequence language model for protein-protein interaction modeling
Liu, J., Chen, H., & Zhang, Y. (2026) · Nature Communications · doi:10.1038/s41467-026-70457-5 · Cited by 8
(2025) · doi:10.1101/2025.07.07.663595 · Cited by 7
Multimodal graph, surface, and language-based model for protein protein interaction prediction
Arteaga, D., Chervov, N., & Poptsova, M. (2026) · Scientific Reports · doi:10.1038/s41598-025-34758-x · Cited by 3
(2026) · doi:10.1073/pnas.2524201123 · Cited by 2
Metadata
- Authors
- Ko, Y. S., Parkinson, J., Liu, C., & Wang, W.
- Venue
- Briefings in Bioinformatics
- Publisher
- Oxford University Press (OUP)
- Published
- 2024-07-25
- Volume
- 25
- Issue
- 5
- ISSN
- 1467-5463 (ISSN-L)
- 1477-4054
- OpenAlex
- W4400984795
- Cited by
- 51
- References
- 10
- FWCI
- 5.93
- Type
- journal article
- Language
- en
- Open access
- gold
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.