Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt
In-silico prediction of protein biophysical traits is often hindered by the limited availability of experimental data and their heterogeneity. Training on limited data can lead to overfitting and poor generalizability to sequences distant from those in the training set. Additionally, inadequate use...
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Taylor & Francis Group
2025-12-01
|
| Series: | mAbs |
| Subjects: | |
| Online Access: | https://www.tandfonline.com/doi/10.1080/19420862.2024.2442750 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1841554605558726656 |
|---|---|
| author | Aubin Ramon Mingyang Ni Olga Predeina Rebecca Gaffey Patrick Kunz Shimobi Onuoha Pietro Sormanni |
| author_facet | Aubin Ramon Mingyang Ni Olga Predeina Rebecca Gaffey Patrick Kunz Shimobi Onuoha Pietro Sormanni |
| author_sort | Aubin Ramon |
| collection | DOAJ |
| description | In-silico prediction of protein biophysical traits is often hindered by the limited availability of experimental data and their heterogeneity. Training on limited data can lead to overfitting and poor generalizability to sequences distant from those in the training set. Additionally, inadequate use of scarce and disparate data can introduce biases during evaluation, leading to unreliable model performances being reported. Here, we present a comprehensive study exploring various approaches for protein fitness prediction from limited data, leveraging pre-trained embeddings, repeated stratified nested cross-validation, and ensemble learning to ensure an unbiased assessment of the performances. We applied our framework to introduce NanoMelt, a predictor of nanobody thermostability trained with a dataset of 640 measurements of apparent melting temperature, obtained by integrating data from the literature with 129 new measurements from this study. We find that an ensemble model stacking multiple regression using diverse sequence embeddings achieves state-of-the-art accuracy in predicting nanobody thermostability. We further demonstrate NanoMelt’s potential to streamline nanobody development by guiding the selection of highly stable nanobodies. We make the curated dataset of nanobody thermostability freely available and NanoMelt accessible as a downloadable software and webserver. |
| format | Article |
| id | doaj-art-0c75e6a1f0aa403d84df18683f8a0d2d |
| institution | Kabale University |
| issn | 1942-0862 1942-0870 |
| language | English |
| publishDate | 2025-12-01 |
| publisher | Taylor & Francis Group |
| record_format | Article |
| series | mAbs |
| spelling | doaj-art-0c75e6a1f0aa403d84df18683f8a0d2d2025-01-08T12:45:19ZengTaylor & Francis GroupmAbs1942-08621942-08702025-12-0117110.1080/19420862.2024.2442750Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMeltAubin Ramon0Mingyang Ni1Olga Predeina2Rebecca Gaffey3Patrick Kunz4Shimobi Onuoha5Pietro Sormanni6Centre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UKCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UKCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UKCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UKDivision of Functional Genome Analysis, German Cancer Research Center (DKFZ), Heidelberg, GermanyChimeris UK, The Works, Cambridge, UKCentre for Misfolding Diseases, Yusuf Hamied Department of Chemistry, University of Cambridge, Cambridge, UKIn-silico prediction of protein biophysical traits is often hindered by the limited availability of experimental data and their heterogeneity. Training on limited data can lead to overfitting and poor generalizability to sequences distant from those in the training set. Additionally, inadequate use of scarce and disparate data can introduce biases during evaluation, leading to unreliable model performances being reported. Here, we present a comprehensive study exploring various approaches for protein fitness prediction from limited data, leveraging pre-trained embeddings, repeated stratified nested cross-validation, and ensemble learning to ensure an unbiased assessment of the performances. We applied our framework to introduce NanoMelt, a predictor of nanobody thermostability trained with a dataset of 640 measurements of apparent melting temperature, obtained by integrating data from the literature with 129 new measurements from this study. We find that an ensemble model stacking multiple regression using diverse sequence embeddings achieves state-of-the-art accuracy in predicting nanobody thermostability. We further demonstrate NanoMelt’s potential to streamline nanobody development by guiding the selection of highly stable nanobodies. We make the curated dataset of nanobody thermostability freely available and NanoMelt accessible as a downloadable software and webserver.https://www.tandfonline.com/doi/10.1080/19420862.2024.2442750Protein fitnessthermostabilityantibody engineeringantibody designnanobodymachine learning |
| spellingShingle | Aubin Ramon Mingyang Ni Olga Predeina Rebecca Gaffey Patrick Kunz Shimobi Onuoha Pietro Sormanni Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt mAbs Protein fitness thermostability antibody engineering antibody design nanobody machine learning |
| title | Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt |
| title_full | Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt |
| title_fullStr | Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt |
| title_full_unstemmed | Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt |
| title_short | Prediction of protein biophysical traits from limited data: a case study on nanobody thermostability through NanoMelt |
| title_sort | prediction of protein biophysical traits from limited data a case study on nanobody thermostability through nanomelt |
| topic | Protein fitness thermostability antibody engineering antibody design nanobody machine learning |
| url | https://www.tandfonline.com/doi/10.1080/19420862.2024.2442750 |
| work_keys_str_mv | AT aubinramon predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt AT mingyangni predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt AT olgapredeina predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt AT rebeccagaffey predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt AT patrickkunz predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt AT shimobionuoha predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt AT pietrosormanni predictionofproteinbiophysicaltraitsfromlimiteddataacasestudyonnanobodythermostabilitythroughnanomelt |