CoChat
Sign Up
PreprintCited by 9Machine Learning in BioinformaticsMicrobial Metabolic Engineering and BioproductionProtein Structure and Dynamics

The DOI matches its Crossref record; 1 of the 1 references it cites were found in a registry or an index.

RP3Net: a deep learning model for predicting recombinant protein production in Escherichia coli

Open access
Tankhilevich, Evgeny, Martinez Cuesta, Sergio, Barrett, Ian, Berg, Carolina, Holmberg Schiavone, Lovisa, Leach, Andrew R·2025·Bioinformatics
DOI 10.1101/2025.05.13.652824PMC PMC12857573

Preprint — published as 10.1093/bioinformatics/btag003

Open this paper in CoChat — free account

Abstract

Recombinant protein expression can be a limiting step in the production of protein reagents for drug discovery and other biotechnology applications. We introduce RP3Net (Recombinant Protein Production Prediction Network), an AI model of small-scale heterologous soluble protein expression in Escherichia coli . RP3Net utilizes the most recent protein and genomic foundational models. A curated dataset of internal experimental results from AstraZeneca (AZ) and publicly available data from the Structural Genomics Consortium (SGC) was used for training, validation and testing of RP3Net. Set Transformer Pooling (STP) aggregation and Meta Label Correction (MLC) with large scale purification data enabled RP3Net to improve Area Under Receiver Operator Curve (AUROC) by 0.15, compared to the baseline model. When experimentally validated on an independent, manually selected set of 97 constructs, RP3Net outperformed currently available models, with an AUROC of 0.83, delivering accurate predictions in 77% of the cases, and correctly identifying successfully expressing constructs in 92% of cases.

Overview

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1101/2025.05.13.652824

Cites

1 of 1 references verified0 citation errors · 0 not checkable

Statistics

Verification

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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

Flagged first, then by citations

Research works printed but not linked: 81

  1. Zanders, E. D. The Science and Business of Drug Discovery: Demystifying the Jargon: Second Edition. The Science and Business of Drug Discovery: Demystifying the Jargon: Second Edition (Springer International Publishing, 2020). · DOI
  2. 10.3389/fddsv.2023.1201419 · DOI
  3. 10.1038/nmeth.f.202 · DOI
  4. 10.1016/j.pep.2007.11.008 · DOI
  5. 10.1016/j.pep.2008.01.008 · DOI
  6. 10.1016/j.pep.2008.07.007 · DOI
  7. 10.1016/j.pep.2008.07.005 · DOI
  8. 10.1016/j.pep.2008.10.022 · DOI
  9. 10.1186/1472-6750-9-21 · DOI
  10. 10.1002/0471140864.ps0524s61 · DOI
  11. 10.1371/journal.pone.0139695 · DOI
  12. 10.1007/978-1-4939-6887-9_2 · DOI
  13. 10.1186/s12934-019-1083-3 · DOI
  14. 10.1007/978-1-0716-0892-0_3 · DOI
  15. 10.1007/978-1-0716-0892-0_4 · DOI
  16. 10.1007/978-1-0716-0892-0_7 · DOI
  17. 10.1007/978-1-0716-0892-0_5 · DOI
  18. 10.1007/978-1-0716-0892-0_6 · DOI
  19. 10.1038/s41598-021-04080-3 · DOI
  20. 10.1371/journal.pone.0271403 · DOI
  21. 10.1007/s00253-023-12805-9 · DOI
  22. 10.1016/B978-0-12-381274-2.00002-9 · DOI
  23. 10.1038/s41467-023-43650-z · DOI
  24. 10.1016/0167-7799(85)90068-X · DOI
  25. 10.1126/science.ade2574 · DOI · Paper page
  26. 10.1073/pnas.2016239118 · DOI · Paper page
  27. 10.1109/TPAMI.2021.3095381 · DOI · Paper page
  28. 10.1093/bioinformatics/btac020 · DOI · Paper page
  29. Nguyen, E. , et al. HyenaDNA: Long-Range Genomic Sequence Modeling at Single Nucleotide Resolution. ArXiv (2023).
  30. 10.1093/bioinformatics/btab083 · DOI · Paper page
  31. Zhou, Z. et al. DNABERT-2: Efficient Foundation Model and Benchmark For Multi-Species Genome. ArXiv (2023).
  32. 10.1038/s42256-024-00791-0 · DOI
  33. 10.1093/bioinformatics/btu739 · DOI · Paper page
  34. 10.1093/NAR/GKAC1052 · DOI · Paper page
  35. 10.1093/nar/gks1195 · DOI · Paper page
  36. 10.1093/nar/gkad903 · DOI
  37. 10.1093/nar/gkac1080 · DOI
  38. 2018, BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding, Proceedings of the Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, 1, 4171
  39. 10.1101/2024.07.01.600583 · DOI
  40. 10.48550/arxiv.1706.03762 · DOI · Paper page
  41. Rao, R. et al. Evaluating Protein Transfer Learning with TAPE. in Advances in Neural Information Processing Systems 9686–9698 (2019). · DOI · Paper page
  42. Dallago, C. , et al. FLIP: Benchmark tasks in fitness landscape inference for proteins. in Thirty-fifth Conference on Neural Information Processing Systems Datasets and Benchmarks Track (Round 2) (2021). · DOI
  43. 10.1093/bioinformatics/btab801 · DOI
  44. Wang, S. & Zhao, H . SADeepcry: a deep learning framework for protein crystallization propensity prediction using self-attention and auto-encoder networks. Brief Bioinform 23, (2022). · DOI
  45. 10.1007/s12539-024-00639-6 · DOI
  46. Li, F.-Z. , Amini, A. P. , Yue, Y. , Yang, K. K. & Lu, A. X. Feature Reuse and Scaling: Understanding Transfer Learning with Protein Language Models. in Proceedings of the 41st International Conference on Machine Learning (eds. Salakhutdinov, R. et al.) vol. 235 27351–27375 (PMLR, 2024). · DOI · Paper page
  47. 10.1007/s10969-011-9106-2 · DOI
  48. Protein production data from the SGC. Preprint at https://www.ebi.ac.uk/biostudies/SGC/studies/S-BSST681 (2022).
  49. 10.1016/j.jsb.2010.06.008 · DOI
  50. 10.1093/nar/28.1.235 · DOI · Paper page
  51. 10.1093/nar/gkac1077 · DOI
  52. 10.1093/bioinformatics/btr229 · DOI
  53. 10.1371/journal.pone.0105902 · DOI
  54. 10.1038/srep21383 · DOI
  55. Zhu, Y.-H. et al. Accurate multistage prediction of protein crystallization propensity using deep-cascade forest with sequence-based features. Brief Bioinform 22, (2021). · DOI
  56. 10.1093/bioinformatics/btaa1102 · DOI
  57. Friedman, J. H . Greedy function approximation: A gradient boosting machine. The Annals of Statistics 29, (2001). · DOI · Paper page
  58. 10.1016/j.jmb.2014.09.026 · DOI · Paper page
  59. 10.1038/s41598-016-0028-x · DOI · Paper page
  60. 10.1093/nar/gkae381 · DOI
  61. Zhang, X. et al. PLM_Sol: predicting protein solubility by benchmarking multiple protein language models with the updated Escherichia coli protein solubility dataset. Brief Bioinform 25, (2024). · DOI
  62. 10.1002/bab.2600 · DOI
  63. 10.3389/fbioe.2022.856049 · DOI
  64. 2022, Advances in Neural Information Processing Systems, 35, 19746
  65. 2019, Proceedings of the 36th International Conference on Machine Learning, 97, 3744
  66. Taraday, M. K. & Baskin, C. Enhanced Meta Label Correction for Coping with Label Corruption. ArXiv (2023). · DOI
  67. Zheng, G. , Awadallah, A. H. & Dumais, S. Meta Label Correction for Noisy Label Learning. in Proceedings of the AAAI Conference on Artificial Intelligence (AAAI) (2021). · DOI
  68. 10.1038/sdata.2016.18 · DOI · Paper page
  69. 10.1126/science.ads0018 · DOI · Paper page
  70. 10.1038/nbt.3988 · DOI · Paper page
  71. Sankar, K. , et al. A Descriptor Set for Quantitative Structure-property Relationship Prediction in Biologics. Mol Inform 41, (2022). · DOI
  72. 10.1016/j.str.2003.10.002 · DOI
  73. 10.1093/nar/gkw306 · DOI
  74. 2016, Uniclust databases of clustered and deeply annotated protein sequences and alignments, Nucleic Acids Res, 45, D170
  75. Paszke, A. et al. PyTorch: An Imperative Style, High-Performance Deep Learning Library. in Advances in Neural Information Processing Systems 32 (Curran Associates, Inc., 2019).
  76. 10.18653/V1/2020.EMNLP-DEMOS.6 · DOI · Paper page
  77. Falcon, W. & The PyTorch Lightning team. PyTorch Lightning. Preprint at (2019).
  78. 10.1109/ASRU57964.2023.10389632 · DOI
  79. Mangrulkar, S. , et al. PEFT: State-of-the-art Parameter-Efficient Fine-Tuning methods. Preprint at https://github.com/huggingface/peft (2022).
  80. 10.1093/nar/gkad1004 · DOI
  81. 10.1093/nar/gkac1046 · DOI · Paper page

Metadata

Authors
Tankhilevich, E., Martinez Cuesta, S., Barrett, I., Berg, C., Holmberg Schiavone, L., & Leach, A. R.
Affiliations
  • European Bioinformatics Institute (GB)
  • Data Sciences and Quantitative Biology, Discovery Sciences, BioPharmaceuticals R&D, Astra-Zeneca, Cambridge, UK
  • AstraZeneca (United Kingdom) (GB)
  • Protein Science, Structure and Biophysics, Discovery Sciences, BioPharmaceuticals R&D, AstraZeneca, Gothenburg, Sweden
Corresponding author
Tankhilevich, Evgeny
ORCID
Venue
Bioinformatics
Publisher
openRxiv
Published
2025-05-16
Volume
42
Issue
1
ISSN
1367-4803 (ISSN-L)
DOI
10.1101/2025.05.13.652824
PMC
PMC12857573
OpenAlex
W7123362149
Cited by
9
References
1
Citations by year
2026: 7 · 2025: 2
FWCI
14.47
Type
journal article
Language
en
Open access
Licence
Locations
  • Bioinformatics (publishedVersion, cc-by)
  • bioRxiv (Cold Spring Harbor Laboratory) (acceptedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
Funders
  • AstraZeneca
  • European Bioinformatics Institute
  • European Molecular Biology Laboratory
Topics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Machine Learning in Bioinformatics
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Microbial Metabolic Engineering and Bioproduction
  • Life Sciences › Biochemistry, Genetics and Molecular Biology › Molecular Biology › Protein Structure and Dynamics
Keywords
recombinant protein expression, Escherichia coli, heterologous expression, deep learning, AstraZeneca
MeSH terms
Deep Learning, Escherichia coli, Recombinant Proteins, Software, Computational Biology

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