Machine learning prediction of methionine and tryptophan photooxidation susceptibility

Photooxidation of methionine (Met) and tryptophan (Trp) residues is common and includes major degradation pathways that often pose a serious threat to the success of therapeutic proteins. Oxidation impacts all steps of protein production, manufacturing, and shelf life. Prediction of oxidation liabil...

Full description

Bibliographic Details
Main Authors: Jared A. Delmar, Eugen Buehler, Ashwin K. Chetty, Agastya Das, Guillermo Miro Quesada, Jihong Wang, Xiaoyu Chen
Format: Article
Language:English
Published: Elsevier 2021-06-01
Series:Molecular Therapy: Methods & Clinical Development
Subjects:
Online Access:http://www.sciencedirect.com/science/article/pii/S2329050121000632
id doaj-5ce2900605144a59996010831a42b91a
record_format Article
spelling doaj-5ce2900605144a59996010831a42b91a2021-06-13T04:38:52ZengElsevierMolecular Therapy: Methods & Clinical Development2329-05012021-06-0121466477Machine learning prediction of methionine and tryptophan photooxidation susceptibilityJared A. Delmar0Eugen Buehler1Ashwin K. Chetty2Agastya Das3Guillermo Miro Quesada4Jihong Wang5Xiaoyu Chen6Biopharmaceuticals Development, R&D, AstraZeneca, Gaithersburg, MD 20878, USA; Corresponding author: Jared A. Delmar, Biopharmaceuticals Development, R&D, AstraZeneca, Gaithersburg, MD 20878, USA.Data Sciences and AI, R&D, AstraZeneca, Gaithersburg, MD 20878, USADepartment of Molecular Biophysics and Biochemistry, Yale University, New Haven, CT 06520, USAKhoury College of Computer Sciences, Northeastern University, Boston, MA 02115, USABiopharmaceuticals Development, R&D, AstraZeneca, Gaithersburg, MD 20878, USABiopharmaceuticals Development, R&D, AstraZeneca, Gaithersburg, MD 20878, USABiopharmaceuticals Development, R&D, AstraZeneca, Gaithersburg, MD 20878, USAPhotooxidation of methionine (Met) and tryptophan (Trp) residues is common and includes major degradation pathways that often pose a serious threat to the success of therapeutic proteins. Oxidation impacts all steps of protein production, manufacturing, and shelf life. Prediction of oxidation liability as early as possible in development is important because many more candidate drugs are discovered than can be tested experimentally. Undetected oxidation liabilities necessitate expensive and time-consuming remediation strategies in development and may lead to good drugs reaching patients slowly. Conversely, sites mischaracterized as oxidation liabilities could result in overengineering and lead to good drugs never reaching patients. To our knowledge, no predictive model for photooxidation of Met or Trp is currently available. We applied the random forest machine learning algorithm to in-house liquid chromatography-tandem mass spectrometry (LC-MS/MS) datasets (Met, n = 421; Trp, n = 342) of tryptic therapeutic protein peptides to create computational models for Met and Trp photooxidation. We show that our machine learning models predict Met and Trp photooxidation likelihood with 0.926 and 0.860 area under the curve (AUC), respectively, and Met photooxidation rate with a correlation coefficient (Q2) of 0.511 and root-mean-square error (RMSE) of 10.9%. We further identify important physical, chemical, and formulation parameters that influence photooxidation. Improvement of biopharmaceutical liability predictions will result in better, more stable drugs, increasing development throughput, product quality, and likelihood of clinical success.http://www.sciencedirect.com/science/article/pii/S2329050121000632oxidationphotostabilitymachine learningpredictiondevelopabilitystability
collection DOAJ
language English
format Article
sources DOAJ
author Jared A. Delmar
Eugen Buehler
Ashwin K. Chetty
Agastya Das
Guillermo Miro Quesada
Jihong Wang
Xiaoyu Chen
spellingShingle Jared A. Delmar
Eugen Buehler
Ashwin K. Chetty
Agastya Das
Guillermo Miro Quesada
Jihong Wang
Xiaoyu Chen
Machine learning prediction of methionine and tryptophan photooxidation susceptibility
Molecular Therapy: Methods & Clinical Development
oxidation
photostability
machine learning
prediction
developability
stability
author_facet Jared A. Delmar
Eugen Buehler
Ashwin K. Chetty
Agastya Das
Guillermo Miro Quesada
Jihong Wang
Xiaoyu Chen
author_sort Jared A. Delmar
title Machine learning prediction of methionine and tryptophan photooxidation susceptibility
title_short Machine learning prediction of methionine and tryptophan photooxidation susceptibility
title_full Machine learning prediction of methionine and tryptophan photooxidation susceptibility
title_fullStr Machine learning prediction of methionine and tryptophan photooxidation susceptibility
title_full_unstemmed Machine learning prediction of methionine and tryptophan photooxidation susceptibility
title_sort machine learning prediction of methionine and tryptophan photooxidation susceptibility
publisher Elsevier
series Molecular Therapy: Methods & Clinical Development
issn 2329-0501
publishDate 2021-06-01
description Photooxidation of methionine (Met) and tryptophan (Trp) residues is common and includes major degradation pathways that often pose a serious threat to the success of therapeutic proteins. Oxidation impacts all steps of protein production, manufacturing, and shelf life. Prediction of oxidation liability as early as possible in development is important because many more candidate drugs are discovered than can be tested experimentally. Undetected oxidation liabilities necessitate expensive and time-consuming remediation strategies in development and may lead to good drugs reaching patients slowly. Conversely, sites mischaracterized as oxidation liabilities could result in overengineering and lead to good drugs never reaching patients. To our knowledge, no predictive model for photooxidation of Met or Trp is currently available. We applied the random forest machine learning algorithm to in-house liquid chromatography-tandem mass spectrometry (LC-MS/MS) datasets (Met, n = 421; Trp, n = 342) of tryptic therapeutic protein peptides to create computational models for Met and Trp photooxidation. We show that our machine learning models predict Met and Trp photooxidation likelihood with 0.926 and 0.860 area under the curve (AUC), respectively, and Met photooxidation rate with a correlation coefficient (Q2) of 0.511 and root-mean-square error (RMSE) of 10.9%. We further identify important physical, chemical, and formulation parameters that influence photooxidation. Improvement of biopharmaceutical liability predictions will result in better, more stable drugs, increasing development throughput, product quality, and likelihood of clinical success.
topic oxidation
photostability
machine learning
prediction
developability
stability
url http://www.sciencedirect.com/science/article/pii/S2329050121000632
work_keys_str_mv AT jaredadelmar machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
AT eugenbuehler machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
AT ashwinkchetty machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
AT agastyadas machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
AT guillermomiroquesada machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
AT jihongwang machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
AT xiaoyuchen machinelearningpredictionofmethionineandtryptophanphotooxidationsusceptibility
_version_ 1721380488350793728