MultiGML: Multimodal graph machine learning for prediction of adverse drug events
Adverse drug events constitute a major challenge for the success of clinical trials. Several computational strategies have been suggested to estimate the risk of adverse drug events in preclinical drug development. While these approaches have demonstrated high utility in practice, they are at the sa...
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2023-09-01
|
| Series: | Heliyon |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2405844023066495 |
| _version_ | 1797669832372518912 |
|---|---|
| author | Sophia Krix Lauren Nicole DeLong Sumit Madan Daniel Domingo-Fernández Ashar Ahmad Sheraz Gul Andrea Zaliani Holger Fröhlich |
| author_facet | Sophia Krix Lauren Nicole DeLong Sumit Madan Daniel Domingo-Fernández Ashar Ahmad Sheraz Gul Andrea Zaliani Holger Fröhlich |
| author_sort | Sophia Krix |
| collection | DOAJ |
| description | Adverse drug events constitute a major challenge for the success of clinical trials. Several computational strategies have been suggested to estimate the risk of adverse drug events in preclinical drug development. While these approaches have demonstrated high utility in practice, they are at the same time limited to specific information sources. Thus, many current computational approaches neglect a wealth of information which results from the integration of different data sources, such as biological protein function, gene expression, chemical compound structure, cell-based imaging and others. In this work we propose an integrative and explainable multi-modal Graph Machine Learning approach (MultiGML), which fuses knowledge graphs with multiple further data modalities to predict drug related adverse events and general drug target-phenotype associations. MultiGML demonstrates excellent prediction performance compared to alternative algorithms, including various traditional knowledge graph embedding techniques. MultiGML distinguishes itself from alternative techniques by providing in-depth explanations of model predictions, which point towards biological mechanisms associated with predictions of an adverse drug event. Hence, MultiGML could be a versatile tool to support decision making in preclinical drug development. |
| first_indexed | 2024-03-11T20:50:31Z |
| format | Article |
| id | doaj.art-4ab39dda7cea48bfb94108aed26f2ee7 |
| institution | Directory Open Access Journal |
| issn | 2405-8440 |
| language | English |
| last_indexed | 2024-03-11T20:50:31Z |
| publishDate | 2023-09-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Heliyon |
| spelling | doaj.art-4ab39dda7cea48bfb94108aed26f2ee72023-10-01T05:59:39ZengElsevierHeliyon2405-84402023-09-0199e19441MultiGML: Multimodal graph machine learning for prediction of adverse drug eventsSophia Krix0Lauren Nicole DeLong1Sumit Madan2Daniel Domingo-Fernández3Ashar Ahmad4Sheraz Gul5Andrea Zaliani6Holger Fröhlich7Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, 53757, Sankt Augustin, Germany; Bonn-Aachen International Center for Information Technology (B-IT), University of Bonn, 53115, Bonn, Germany; Fraunhofer Center for Machine Learning, GermanyDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, 53757, Sankt Augustin, Germany; Artificial Intelligence and its Applications Institute, School of Informatics, University of Edinburgh, 10 Crichton Street, EH8 9AB, UKDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, 53757, Sankt Augustin, Germany; Department of Computer Science, University of Bonn, 53115, Bonn, GermanyDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, 53757, Sankt Augustin, Germany; Fraunhofer Center for Machine Learning, Germany; Enveda Biosciences, Boulder, CO, 80301, USABonn-Aachen International Center for Information Technology (B-IT), University of Bonn, 53115, Bonn, Germany; Grunenthal GmbH, 52099, Aachen, GermanyFraunhofer Institute for Translational Medicine and Pharmacology ITMP, Schnackenburgallee 114, 22525, Hamburg, Germany; Fraunhofer Cluster of Excellence for Immune-Mediated Diseases CIMD, Schnackenburgallee 114, 22525, Hamburg, GermanyFraunhofer Institute for Translational Medicine and Pharmacology ITMP, Schnackenburgallee 114, 22525, Hamburg, Germany; Fraunhofer Cluster of Excellence for Immune-Mediated Diseases CIMD, Schnackenburgallee 114, 22525, Hamburg, GermanyDepartment of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, 53757, Sankt Augustin, Germany; Bonn-Aachen International Center for Information Technology (B-IT), University of Bonn, 53115, Bonn, Germany; Corresponding author. Department of Bioinformatics, Fraunhofer Institute for Algorithms and Scientific Computing (SCAI), Schloss Birlinghoven, 53757, Sankt Augustin, Germany.Adverse drug events constitute a major challenge for the success of clinical trials. Several computational strategies have been suggested to estimate the risk of adverse drug events in preclinical drug development. While these approaches have demonstrated high utility in practice, they are at the same time limited to specific information sources. Thus, many current computational approaches neglect a wealth of information which results from the integration of different data sources, such as biological protein function, gene expression, chemical compound structure, cell-based imaging and others. In this work we propose an integrative and explainable multi-modal Graph Machine Learning approach (MultiGML), which fuses knowledge graphs with multiple further data modalities to predict drug related adverse events and general drug target-phenotype associations. MultiGML demonstrates excellent prediction performance compared to alternative algorithms, including various traditional knowledge graph embedding techniques. MultiGML distinguishes itself from alternative techniques by providing in-depth explanations of model predictions, which point towards biological mechanisms associated with predictions of an adverse drug event. Hence, MultiGML could be a versatile tool to support decision making in preclinical drug development.http://www.sciencedirect.com/science/article/pii/S2405844023066495Machine learningKnowledge graphAdverse eventGraph neural networkGraph attention networkGraph convolutional network |
| spellingShingle | Sophia Krix Lauren Nicole DeLong Sumit Madan Daniel Domingo-Fernández Ashar Ahmad Sheraz Gul Andrea Zaliani Holger Fröhlich MultiGML: Multimodal graph machine learning for prediction of adverse drug events Heliyon Machine learning Knowledge graph Adverse event Graph neural network Graph attention network Graph convolutional network |
| title | MultiGML: Multimodal graph machine learning for prediction of adverse drug events |
| title_full | MultiGML: Multimodal graph machine learning for prediction of adverse drug events |
| title_fullStr | MultiGML: Multimodal graph machine learning for prediction of adverse drug events |
| title_full_unstemmed | MultiGML: Multimodal graph machine learning for prediction of adverse drug events |
| title_short | MultiGML: Multimodal graph machine learning for prediction of adverse drug events |
| title_sort | multigml multimodal graph machine learning for prediction of adverse drug events |
| topic | Machine learning Knowledge graph Adverse event Graph neural network Graph attention network Graph convolutional network |
| url | http://www.sciencedirect.com/science/article/pii/S2405844023066495 |
| work_keys_str_mv | AT sophiakrix multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT laurennicoledelong multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT sumitmadan multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT danieldomingofernandez multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT asharahmad multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT sherazgul multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT andreazaliani multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents AT holgerfrohlich multigmlmultimodalgraphmachinelearningforpredictionofadversedrugevents |