Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures

Abstract The utility of pathway signatures lies in their capability to determine whether a specific pathway or biological process is dysregulated in a given patient. These signatures have been widely used in machine learning (ML) methods for a variety of applications including precision medicine, dr...

Full description

Bibliographic Details
Main Authors: Sepehr Golriz Khatami, Sarah Mubeen, Vinay Srinivas Bharadhwaj, Alpha Tom Kodamullil, Martin Hofmann-Apitius, Daniel Domingo-Fernández
Format: Article
Language:English
Published: Nature Portfolio 2021-10-01
Series:npj Systems Biology and Applications
Online Access:https://doi.org/10.1038/s41540-021-00199-1
_version_ 1819044854759424000
author Sepehr Golriz Khatami
Sarah Mubeen
Vinay Srinivas Bharadhwaj
Alpha Tom Kodamullil
Martin Hofmann-Apitius
Daniel Domingo-Fernández
author_facet Sepehr Golriz Khatami
Sarah Mubeen
Vinay Srinivas Bharadhwaj
Alpha Tom Kodamullil
Martin Hofmann-Apitius
Daniel Domingo-Fernández
author_sort Sepehr Golriz Khatami
collection DOAJ
description Abstract The utility of pathway signatures lies in their capability to determine whether a specific pathway or biological process is dysregulated in a given patient. These signatures have been widely used in machine learning (ML) methods for a variety of applications including precision medicine, drug repurposing, and drug discovery. In this work, we leverage highly predictive ML models for drug response simulation in individual patients by calibrating the pathway activity scores of disease samples. Using these ML models and an intuitive scoring algorithm to modify the signatures of patients, we evaluate whether a given sample that was formerly classified as diseased, could be predicted as normal following drug treatment simulation. We then use this technique as a proxy for the identification of potential drug candidates. Furthermore, we demonstrate the ability of our methodology to successfully identify approved and clinically investigated drugs for four different cancers, outperforming six comparable state-of-the-art methods. We also show how this approach can deconvolute a drugs’ mechanism of action and propose combination therapies. Taken together, our methodology could be promising to support clinical decision-making in personalized medicine by simulating a drugs’ effect on a given patient.
first_indexed 2024-12-21T10:19:18Z
format Article
id doaj.art-6393f011511e4fadb3e36678d182282d
institution Directory Open Access Journal
issn 2056-7189
language English
last_indexed 2024-12-21T10:19:18Z
publishDate 2021-10-01
publisher Nature Portfolio
record_format Article
series npj Systems Biology and Applications
spelling doaj.art-6393f011511e4fadb3e36678d182282d2022-12-21T19:07:30ZengNature Portfolionpj Systems Biology and Applications2056-71892021-10-01711910.1038/s41540-021-00199-1Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signaturesSepehr Golriz Khatami0Sarah Mubeen1Vinay Srinivas Bharadhwaj2Alpha Tom Kodamullil3Martin Hofmann-Apitius4Daniel Domingo-Fernández5Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific ComputingDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific ComputingDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific ComputingDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific ComputingDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific ComputingDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific ComputingAbstract The utility of pathway signatures lies in their capability to determine whether a specific pathway or biological process is dysregulated in a given patient. These signatures have been widely used in machine learning (ML) methods for a variety of applications including precision medicine, drug repurposing, and drug discovery. In this work, we leverage highly predictive ML models for drug response simulation in individual patients by calibrating the pathway activity scores of disease samples. Using these ML models and an intuitive scoring algorithm to modify the signatures of patients, we evaluate whether a given sample that was formerly classified as diseased, could be predicted as normal following drug treatment simulation. We then use this technique as a proxy for the identification of potential drug candidates. Furthermore, we demonstrate the ability of our methodology to successfully identify approved and clinically investigated drugs for four different cancers, outperforming six comparable state-of-the-art methods. We also show how this approach can deconvolute a drugs’ mechanism of action and propose combination therapies. Taken together, our methodology could be promising to support clinical decision-making in personalized medicine by simulating a drugs’ effect on a given patient.https://doi.org/10.1038/s41540-021-00199-1
spellingShingle Sepehr Golriz Khatami
Sarah Mubeen
Vinay Srinivas Bharadhwaj
Alpha Tom Kodamullil
Martin Hofmann-Apitius
Daniel Domingo-Fernández
Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures
npj Systems Biology and Applications
title Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures
title_full Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures
title_fullStr Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures
title_full_unstemmed Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures
title_short Using predictive machine learning models for drug response simulation by calibrating patient-specific pathway signatures
title_sort using predictive machine learning models for drug response simulation by calibrating patient specific pathway signatures
url https://doi.org/10.1038/s41540-021-00199-1
work_keys_str_mv AT sepehrgolrizkhatami usingpredictivemachinelearningmodelsfordrugresponsesimulationbycalibratingpatientspecificpathwaysignatures
AT sarahmubeen usingpredictivemachinelearningmodelsfordrugresponsesimulationbycalibratingpatientspecificpathwaysignatures
AT vinaysrinivasbharadhwaj usingpredictivemachinelearningmodelsfordrugresponsesimulationbycalibratingpatientspecificpathwaysignatures
AT alphatomkodamullil usingpredictivemachinelearningmodelsfordrugresponsesimulationbycalibratingpatientspecificpathwaysignatures
AT martinhofmannapitius usingpredictivemachinelearningmodelsfordrugresponsesimulationbycalibratingpatientspecificpathwaysignatures
AT danieldomingofernandez usingpredictivemachinelearningmodelsfordrugresponsesimulationbycalibratingpatientspecificpathwaysignatures