Prediction of aggregation in monoclonal antibodies from molecular surface curvature

Abstract Protein aggregation is one of the key challenges in the biopharmaceutical industry as its control is crucial in achieving long-term stability and efficacy of biopharmaceuticals. Attempts have been made to develop regression models for predicting the aggregation of monoclonal antibodies in s...

Full description

Saved in:
Bibliographic Details
Main Authors: Benjamin Knez, Lara Erzin, Žiga Kos, Drago Kuzman, Miha Ravnik
Format: Article
Language:English
Published: Nature Portfolio 2025-08-01
Series:Scientific Reports
Online Access:https://doi.org/10.1038/s41598-025-13527-w
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849766302757945344
author Benjamin Knez
Lara Erzin
Žiga Kos
Drago Kuzman
Miha Ravnik
author_facet Benjamin Knez
Lara Erzin
Žiga Kos
Drago Kuzman
Miha Ravnik
author_sort Benjamin Knez
collection DOAJ
description Abstract Protein aggregation is one of the key challenges in the biopharmaceutical industry as its control is crucial in achieving long-term stability and efficacy of biopharmaceuticals. Attempts have been made to develop regression models for predicting the aggregation of monoclonal antibodies in solution using machine learning methods. These efforts have yielded varying levels of success, with current state-of-the-art AI approaches achieving good prediction accuracies ( $$r=0.86$$ ). Here, we demonstrate the prediction of aggregation rate in monoclonal antibodies with beyond state-of-the-art reliability using a coupled AI-MD-Molecular surface curvature modelling platform. The scientific novelty of this approach lies in using local geometrical surface curvature of proteins as the core element for protein stability analysis. By combining local surface curvature and hydrophobicity, as derived from time-dependent MD simulations, we are able to construct aggregation predictive features that, when coupled with linear regression machine learning techniques, give a high prediction accuracy ( $$r=0.91$$ ) on a dataset of 20 molecules. More generally, this approach shows significant potential for quantitative in silico screening and prediction of protein aggregation, which is of great scientific and industrial relevance, particularly in biopharmaceutics.
format Article
id doaj-art-e14f5c846dca4c928041c3533e981efa
institution DOAJ
issn 2045-2322
language English
publishDate 2025-08-01
publisher Nature Portfolio
record_format Article
series Scientific Reports
spelling doaj-art-e14f5c846dca4c928041c3533e981efa2025-08-20T03:04:38ZengNature PortfolioScientific Reports2045-23222025-08-0115111110.1038/s41598-025-13527-wPrediction of aggregation in monoclonal antibodies from molecular surface curvatureBenjamin Knez0Lara Erzin1Žiga Kos2Drago Kuzman3Miha Ravnik4Novartis LLCFaculty of Mathematics and Physics, University of LjubljanaFaculty of Mathematics and Physics, University of LjubljanaNovartis LLCFaculty of Mathematics and Physics, University of LjubljanaAbstract Protein aggregation is one of the key challenges in the biopharmaceutical industry as its control is crucial in achieving long-term stability and efficacy of biopharmaceuticals. Attempts have been made to develop regression models for predicting the aggregation of monoclonal antibodies in solution using machine learning methods. These efforts have yielded varying levels of success, with current state-of-the-art AI approaches achieving good prediction accuracies ( $$r=0.86$$ ). Here, we demonstrate the prediction of aggregation rate in monoclonal antibodies with beyond state-of-the-art reliability using a coupled AI-MD-Molecular surface curvature modelling platform. The scientific novelty of this approach lies in using local geometrical surface curvature of proteins as the core element for protein stability analysis. By combining local surface curvature and hydrophobicity, as derived from time-dependent MD simulations, we are able to construct aggregation predictive features that, when coupled with linear regression machine learning techniques, give a high prediction accuracy ( $$r=0.91$$ ) on a dataset of 20 molecules. More generally, this approach shows significant potential for quantitative in silico screening and prediction of protein aggregation, which is of great scientific and industrial relevance, particularly in biopharmaceutics.https://doi.org/10.1038/s41598-025-13527-w
spellingShingle Benjamin Knez
Lara Erzin
Žiga Kos
Drago Kuzman
Miha Ravnik
Prediction of aggregation in monoclonal antibodies from molecular surface curvature
Scientific Reports
title Prediction of aggregation in monoclonal antibodies from molecular surface curvature
title_full Prediction of aggregation in monoclonal antibodies from molecular surface curvature
title_fullStr Prediction of aggregation in monoclonal antibodies from molecular surface curvature
title_full_unstemmed Prediction of aggregation in monoclonal antibodies from molecular surface curvature
title_short Prediction of aggregation in monoclonal antibodies from molecular surface curvature
title_sort prediction of aggregation in monoclonal antibodies from molecular surface curvature
url https://doi.org/10.1038/s41598-025-13527-w
work_keys_str_mv AT benjaminknez predictionofaggregationinmonoclonalantibodiesfrommolecularsurfacecurvature
AT laraerzin predictionofaggregationinmonoclonalantibodiesfrommolecularsurfacecurvature
AT zigakos predictionofaggregationinmonoclonalantibodiesfrommolecularsurfacecurvature
AT dragokuzman predictionofaggregationinmonoclonalantibodiesfrommolecularsurfacecurvature
AT miharavnik predictionofaggregationinmonoclonalantibodiesfrommolecularsurfacecurvature