Leveraging multi-modal feature learning for predictions of antibody viscosity

The shift toward subcutaneous administration for biologic therapeutics has gained momentum due to its patient-friendly nature, convenience, reduced healthcare burden, and improved compliance compared to traditional intravenous infusions. However, a significant challenge associated with this transiti...

Full description

Saved in:
Bibliographic Details
Main Authors: Krishna D. B. Anapindi, Kai Liu, Willie Wang, Yao Yu, Yan He, Edward J. Hsieh, Ying Huang, Daniela Tomazela
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:mAbs
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/19420862.2025.2490788
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850185656103337984
author Krishna D. B. Anapindi
Kai Liu
Willie Wang
Yao Yu
Yan He
Edward J. Hsieh
Ying Huang
Daniela Tomazela
author_facet Krishna D. B. Anapindi
Kai Liu
Willie Wang
Yao Yu
Yan He
Edward J. Hsieh
Ying Huang
Daniela Tomazela
author_sort Krishna D. B. Anapindi
collection DOAJ
description The shift toward subcutaneous administration for biologic therapeutics has gained momentum due to its patient-friendly nature, convenience, reduced healthcare burden, and improved compliance compared to traditional intravenous infusions. However, a significant challenge associated with this transition is managing the viscosity of the administered solutions. High viscosity poses substantial development and manufacturability challenges, directly affecting patients by increasing injection time and causing pain at the injection site. Furthermore, high viscosity formulations can prolong residence time at the injection site, affecting absorption kinetics and potentially altering the intended pharmacological profile and therapeutic efficacy of the biologic candidate. Here, we report the application of a multimodal feature learning workflow for predicting the viscosity of antibodies in therapeutics discovery. It integrates multiple data sources including sequence, structural, physicochemical properties, as well as embeddings from a language model. This approach enables the model to learn from various underlying rules, such as physicochemical rules from molecular simulations and protein evolutionary patterns captured by large, pre-trained deep learning models. By comparing the effectiveness of this approach to other selected published viscosity prediction methods, this study provides insights into their intrinsic viscosity predictive potential and usability in early-stage therapeutics antibody development pipelines.
format Article
id doaj-art-e8eaaea3ed044aafa2f387a1bf031607
institution OA Journals
issn 1942-0862
1942-0870
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series mAbs
spelling doaj-art-e8eaaea3ed044aafa2f387a1bf0316072025-08-20T02:16:39ZengTaylor & Francis GroupmAbs1942-08621942-08702025-12-0117110.1080/19420862.2025.2490788Leveraging multi-modal feature learning for predictions of antibody viscosityKrishna D. B. Anapindi0Kai Liu1Willie Wang2Yao Yu3Yan He4Edward J. Hsieh5Ying Huang6Daniela Tomazela7Protein Therapeutics, Gilead Sciences, Foster City, CA, USAResearch Data Sciences, Gilead Sciences, Foster City, CA, USAProtein Therapeutics, Gilead Sciences, Foster City, CA, USAProtein Therapeutics, Gilead Sciences, Foster City, CA, USAProtein Therapeutics, Gilead Sciences, Foster City, CA, USAProtein Therapeutics, Gilead Sciences, Foster City, CA, USAResearch Data Sciences, Gilead Sciences, Foster City, CA, USAProtein Therapeutics, Gilead Sciences, Foster City, CA, USAThe shift toward subcutaneous administration for biologic therapeutics has gained momentum due to its patient-friendly nature, convenience, reduced healthcare burden, and improved compliance compared to traditional intravenous infusions. However, a significant challenge associated with this transition is managing the viscosity of the administered solutions. High viscosity poses substantial development and manufacturability challenges, directly affecting patients by increasing injection time and causing pain at the injection site. Furthermore, high viscosity formulations can prolong residence time at the injection site, affecting absorption kinetics and potentially altering the intended pharmacological profile and therapeutic efficacy of the biologic candidate. Here, we report the application of a multimodal feature learning workflow for predicting the viscosity of antibodies in therapeutics discovery. It integrates multiple data sources including sequence, structural, physicochemical properties, as well as embeddings from a language model. This approach enables the model to learn from various underlying rules, such as physicochemical rules from molecular simulations and protein evolutionary patterns captured by large, pre-trained deep learning models. By comparing the effectiveness of this approach to other selected published viscosity prediction methods, this study provides insights into their intrinsic viscosity predictive potential and usability in early-stage therapeutics antibody development pipelines.https://www.tandfonline.com/doi/10.1080/19420862.2025.2490788Multimodal feature learningantibody viscosity predictionmachine learningbioinformatics
spellingShingle Krishna D. B. Anapindi
Kai Liu
Willie Wang
Yao Yu
Yan He
Edward J. Hsieh
Ying Huang
Daniela Tomazela
Leveraging multi-modal feature learning for predictions of antibody viscosity
mAbs
Multimodal feature learning
antibody viscosity prediction
machine learning
bioinformatics
title Leveraging multi-modal feature learning for predictions of antibody viscosity
title_full Leveraging multi-modal feature learning for predictions of antibody viscosity
title_fullStr Leveraging multi-modal feature learning for predictions of antibody viscosity
title_full_unstemmed Leveraging multi-modal feature learning for predictions of antibody viscosity
title_short Leveraging multi-modal feature learning for predictions of antibody viscosity
title_sort leveraging multi modal feature learning for predictions of antibody viscosity
topic Multimodal feature learning
antibody viscosity prediction
machine learning
bioinformatics
url https://www.tandfonline.com/doi/10.1080/19420862.2025.2490788
work_keys_str_mv AT krishnadbanapindi leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT kailiu leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT williewang leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT yaoyu leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT yanhe leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT edwardjhsieh leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT yinghuang leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity
AT danielatomazela leveragingmultimodalfeaturelearningforpredictionsofantibodyviscosity