Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning

Highly concentrated antibody solutions are necessary for developing subcutaneous injections but often exhibit high viscosities, posing challenges in antibody-drug development, manufacturing, and administration. Previous computational models were only limited to a few dozen data points for training,...

Full description

Saved in:
Bibliographic Details
Main Authors: Lateefat A. Kalejaye, Jia-Min Chu, I-En Wu, Bismark Amofah, Amber Lee, Mark Hutchinson, Chacko Chakiath, Andrew Dippel, Gilad Kaplan, Melissa Damschroder, Valentin Stanev, Maryam Pouryahya, Mehdi Boroumand, Jenna Caldwell, Alison Hinton, Madison Kreitz, Mitali Shah, Austin Gallegos, Neil Mody, Pin-Kuang Lai
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:mAbs
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/19420862.2025.2483944
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850034231389978624
author Lateefat A. Kalejaye
Jia-Min Chu
I-En Wu
Bismark Amofah
Amber Lee
Mark Hutchinson
Chacko Chakiath
Andrew Dippel
Gilad Kaplan
Melissa Damschroder
Valentin Stanev
Maryam Pouryahya
Mehdi Boroumand
Jenna Caldwell
Alison Hinton
Madison Kreitz
Mitali Shah
Austin Gallegos
Neil Mody
Pin-Kuang Lai
author_facet Lateefat A. Kalejaye
Jia-Min Chu
I-En Wu
Bismark Amofah
Amber Lee
Mark Hutchinson
Chacko Chakiath
Andrew Dippel
Gilad Kaplan
Melissa Damschroder
Valentin Stanev
Maryam Pouryahya
Mehdi Boroumand
Jenna Caldwell
Alison Hinton
Madison Kreitz
Mitali Shah
Austin Gallegos
Neil Mody
Pin-Kuang Lai
author_sort Lateefat A. Kalejaye
collection DOAJ
description Highly concentrated antibody solutions are necessary for developing subcutaneous injections but often exhibit high viscosities, posing challenges in antibody-drug development, manufacturing, and administration. Previous computational models were only limited to a few dozen data points for training, a bottleneck for generalizability. In this study, we measured the viscosity of a panel of 229 monoclonal antibodies (mAbs) to develop predictive models for high concentration mAb screening. We developed DeepViscosity, consisting of 102 ensemble artificial neural network models to classify low-viscosity (≤20 cP) and high-viscosity (>20 cP) mAbs at 150 mg/mL, using 30 features from a sequence-based DeepSP model. Two independent test sets, comprising 16 and 38 mAbs with known experimental viscosity, were used to assess DeepViscosity’s generalizability. The model exhibited an accuracy of 87.5% and 89.5% on both test sets, respectively, surpassing other predictive methods. DeepViscosity will facilitate early-stage antibody development to select low-viscosity antibodies for improved manufacturability and formulation properties, critical for subcutaneous drug delivery. The webserver-based application can be freely accessed via https://devpred.onrender.com/DeepViscosity.
format Article
id doaj-art-e7dcd89959a34066856e99ebdd9e026a
institution DOAJ
issn 1942-0862
1942-0870
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series mAbs
spelling doaj-art-e7dcd89959a34066856e99ebdd9e026a2025-08-20T02:57:53ZengTaylor & Francis GroupmAbs1942-08621942-08702025-12-0117110.1080/19420862.2025.2483944Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learningLateefat A. Kalejaye0Jia-Min Chu1I-En Wu2Bismark Amofah3Amber Lee4Mark Hutchinson5Chacko Chakiath6Andrew Dippel7Gilad Kaplan8Melissa Damschroder9Valentin Stanev10Maryam Pouryahya11Mehdi Boroumand12Jenna Caldwell13Alison Hinton14Madison Kreitz15Mitali Shah16Austin Gallegos17Neil Mody18Pin-Kuang Lai19Department of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken, NJ, USADepartment of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken, NJ, USADepartment of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken, NJ, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USABiologics Engineering, R&D, AstraZeneca, Gaithersburg, MD, USAData Science and Modelling, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USAData Science and Modelling, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USAData Science and Modelling, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADosage Form Design and Development, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADosage Form Design and Development, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADosage Form Design and Development, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADosage Form Design and Development, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADosage Form Design and Development, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADosage Form Design and Development, BioPharmaceuticals R&D, AstraZeneca, Gaithersburg, MD, USADepartment of Chemical Engineering and Materials Science, Stevens Institute of Technology, Hoboken, NJ, USAHighly concentrated antibody solutions are necessary for developing subcutaneous injections but often exhibit high viscosities, posing challenges in antibody-drug development, manufacturing, and administration. Previous computational models were only limited to a few dozen data points for training, a bottleneck for generalizability. In this study, we measured the viscosity of a panel of 229 monoclonal antibodies (mAbs) to develop predictive models for high concentration mAb screening. We developed DeepViscosity, consisting of 102 ensemble artificial neural network models to classify low-viscosity (≤20 cP) and high-viscosity (>20 cP) mAbs at 150 mg/mL, using 30 features from a sequence-based DeepSP model. Two independent test sets, comprising 16 and 38 mAbs with known experimental viscosity, were used to assess DeepViscosity’s generalizability. The model exhibited an accuracy of 87.5% and 89.5% on both test sets, respectively, surpassing other predictive methods. DeepViscosity will facilitate early-stage antibody development to select low-viscosity antibodies for improved manufacturability and formulation properties, critical for subcutaneous drug delivery. The webserver-based application can be freely accessed via https://devpred.onrender.com/DeepViscosity.https://www.tandfonline.com/doi/10.1080/19420862.2025.2483944Antibody viscosityensemble deep learninghigh-concentration formulationsmonoclonal antibodies
spellingShingle Lateefat A. Kalejaye
Jia-Min Chu
I-En Wu
Bismark Amofah
Amber Lee
Mark Hutchinson
Chacko Chakiath
Andrew Dippel
Gilad Kaplan
Melissa Damschroder
Valentin Stanev
Maryam Pouryahya
Mehdi Boroumand
Jenna Caldwell
Alison Hinton
Madison Kreitz
Mitali Shah
Austin Gallegos
Neil Mody
Pin-Kuang Lai
Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning
mAbs
Antibody viscosity
ensemble deep learning
high-concentration formulations
monoclonal antibodies
title Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning
title_full Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning
title_fullStr Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning
title_full_unstemmed Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning
title_short Accelerating high-concentration monoclonal antibody development with large-scale viscosity data and ensemble deep learning
title_sort accelerating high concentration monoclonal antibody development with large scale viscosity data and ensemble deep learning
topic Antibody viscosity
ensemble deep learning
high-concentration formulations
monoclonal antibodies
url https://www.tandfonline.com/doi/10.1080/19420862.2025.2483944
work_keys_str_mv AT lateefatakalejaye acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT jiaminchu acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT ienwu acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT bismarkamofah acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT amberlee acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT markhutchinson acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT chackochakiath acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT andrewdippel acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT giladkaplan acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT melissadamschroder acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT valentinstanev acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT maryampouryahya acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT mehdiboroumand acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT jennacaldwell acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT alisonhinton acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT madisonkreitz acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT mitalishah acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT austingallegos acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT neilmody acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning
AT pinkuanglai acceleratinghighconcentrationmonoclonalantibodydevelopmentwithlargescaleviscositydataandensembledeeplearning