Version 1
: Received: 8 October 2020 / Approved: 9 October 2020 / Online: 9 October 2020 (11:13:19 CEST)
Version 2
: Received: 19 November 2020 / Approved: 20 November 2020 / Online: 20 November 2020 (11:30:03 CET)
How to cite:
Jacobs, I.; Maragoudakis, M. De Novo Drug Design using Artificial Intelligence ASYNT-GAN. Preprints2020, 2020100196. https://doi.org/10.20944/preprints202010.0196.v2
Jacobs, I.; Maragoudakis, M. De Novo Drug Design using Artificial Intelligence ASYNT-GAN. Preprints 2020, 2020100196. https://doi.org/10.20944/preprints202010.0196.v2
Jacobs, I.; Maragoudakis, M. De Novo Drug Design using Artificial Intelligence ASYNT-GAN. Preprints2020, 2020100196. https://doi.org/10.20944/preprints202010.0196.v2
APA Style
Jacobs, I., & Maragoudakis, M. (2020). De Novo Drug Design using Artificial Intelligence ASYNT-GAN. Preprints. https://doi.org/10.20944/preprints202010.0196.v2
Chicago/Turabian Style
Jacobs, I. and Manolis Maragoudakis. 2020 "De Novo Drug Design using Artificial Intelligence ASYNT-GAN" Preprints. https://doi.org/10.20944/preprints202010.0196.v2
Abstract
In this paper we propose the generation of synthetic small and more sophisticated molecule structures that optimize the binding affinity to a target (ASYNT-GAN). To achieve this we leverage on three important achievements in A.I.: Attention, Deep Learning on Graphs and Generative Adversarial Networks. Similar to text generation based on parts of text we are able to generate a molecule architecture based on an existing target. By adopting this approach, we propose a novel way of searching for existing compounds that are suitable candidates. Similar to question and answer Natural Language solutions we are able to find drugs with highest relevance to a target. We are able to identify substructures of the molecular structure that are the most suitable for binding. In addition, we are proposing a novel way of generating the molecule in 3D space in such a way that the binding is optimized. We show that we are able to generate compound structures and protein structures that are optimised for binding to a target.
Keywords
drug discovery; artificial intelligence; protein discovery; binding prediction; synthetic molecule generation; synthetic drug
Subject
Medicine and Pharmacology, Immunology and Allergy
Copyright:
This is an open access article distributed under the Creative Commons Attribution License which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Received:
20 November 2020
Commenter:
Ivan Jacobs
Commenter's Conflict of Interests:
Author
Comment: Response 1: As found on line 30 to 99: We have changed the introduction and added additional information comparing our approach with other currently applied techniques. We have as well added additional information explaining the loss of information in the current approaches. Response 4: As found on line 218 and line 223: The figures 7 and 8 have been redone. However the number of figures have not been reduced. From the other reviews we came to the conclusion that some of the a.i. related topics need additional information. We added a concrete example and added new figures to support the comprehension. As found on line 275 to 337 Response 5: The evaluation metrics and CD are moved to the Method section on line 173 and line 166 .
Commenter: Ivan Jacobs
Commenter's Conflict of Interests: Author
Response 4: As found on line 218 and line 223: The figures 7 and 8 have been redone. However the number of figures have not been reduced. From the other reviews we came to the conclusion that some of the a.i. related topics need additional information. We added a concrete example and added new figures to support the comprehension. As found on line 275 to 337
Response 5: The evaluation metrics and CD are moved to the Method section on line 173 and line 166 .