Filters








15 Hits in 2.8 sec

Mol-CycleGAN: a generative model for molecular optimization

Łukasz Maziarka, Agnieszka Pocha, Jan Kaczmarczyk, Krzysztof Rataj, Tomasz Danel, Michał Warchoł
2020 Journal of Cheminformatics  
To improve the compound design process, we introduce Mol-CycleGAN-a CycleGAN-based model that generates optimized compounds with high structural similarity to the original ones.  ...  Namely, given a molecule our model generates a structurally similar one with an optimized value of the considered property.  ...  Acknowledgements We would like to thank Sabina Podlewska for her helpful comments and for fruitful discussions.  ... 
doi:10.1186/s13321-019-0404-1 pmid:33431006 fatcat:zhocx6tl3bfvvaborcs7sch464

Mol-CycleGAN - a generative model for molecular optimization [article]

Łukasz Maziarka, Agnieszka Pocha, Jan Kaczmarczyk, Krzysztof Rataj, Michał Warchoł
2019 arXiv   pre-print
To augment the compound design process we introduce Mol-CycleGAN - a CycleGAN-based model that generates optimized compounds with high structural similarity to the original ones.  ...  Namely, given a molecule our model generates a structurally similar one with an optimized value of the considered property.  ...  Acknowledgement We would like to thank Sabina Podlewska for her helpful comments and for fruitful discussions.  ... 
arXiv:1902.02119v1 fatcat:bv5ukbl2ivhr7mdigfze7lslbe

Mol-CycleGAN - A Generative Model for Molecular Optimization [chapter]

Łukasz Maziarka, Agnieszka Pocha, Jan Kaczmarczyk, Krzysztof Rataj, Michał Warchoł
2019 Lecture Notes in Computer Science  
To improve this process, we introduce Mol-CycleGAN -a CycleGAN-based model that generates compounds optimized for a selected property, while aiming to retain the already optimized ones.  ...  In the task of constrained optimization of penalized logP of drug-like molecules our model significantly outperforms previous results.  ...  Mol-CycleGAN Mol-CycleGAN is a novel method of performing compound optimization by learning from the sets of molecules with and without the desired molecular property (denoted by the sets X and Y ).  ... 
doi:10.1007/978-3-030-30493-5_77 fatcat:t36xvxezdzbkriycwkjgnojmt4

MERMAID: an open source automated hit-to-lead method based on deep reinforcement learning

Daiki Erikawa, Nobuaki Yasuo, Masakazu Sekijima
2021 Journal of Cheminformatics  
In this study, we have developed a SMILES-based generative model that can be generated starting from a certain molecule.  ...  We validated our method using a molecule dataset obtained from the ZINC database and successfully generated molecules that were both well optimized for the objectives of the quantitative estimate of drug-likeness  ...  Acknowledgements This work was partially supported by the Platform Project for Supporting Drug Discovery and Life Science Research (Basis for Supporting Innovative Drug Discovery and Life Science Research  ... 
doi:10.1186/s13321-021-00572-6 pmid:34838134 pmcid:PMC8626955 fatcat:lwc3z5ttyfbxlpvbpakdnwjwby

Relevant Applications of Generative Adversarial Networks in Drug Design and Discovery: Molecular De Novo Design, Dimensionality Reduction, and De Novo Peptide and Protein Design

Eugene Lin, Chieh-Hsin Lin, Hsien-Yuan Lane
2020 Molecules  
Moreover, we outline the limitations in regard to the previous drug design and discovery studies using GAN models. Finally, we present a discussion of directions and challenges for future research.  ...  In this review, we focus on the latest developments for three particular arenas in drug design and discovery research using deep learning approaches, such as generative adversarial network (GAN) frameworks  ...  [44] implemented a deep learning GAN architecture called the Mol-CycleGAN structure to produce optimized molecular compounds where their molecular structures were highly similar to the original ones  ... 
doi:10.3390/molecules25143250 pmid:32708785 fatcat:rrik322g6vbetaubwjb3rtvajm

From Big Data to Artificial Intelligence: chemoinformatics meets new challenges

Igor V. Tetko, Ola Engkvist
2020 Journal of Cheminformatics  
The increasing volume of biomedical data in chemistry and life sciences requires development of new methods and approaches for their analysis.  ...  The methods developed in these studies are general ones and can be used to enhance other GMs such as scaffold decoration [18] or Mol-CycleGAN [19] .  ...  The second group of articles deals with novel machine-learning algorithms such as the use of generative models (GMs) for molecular de novo design in drug discovery.  ... 
doi:10.1186/s13321-020-00475-y pmid:33339533 fatcat:3bpwkzpzxfbo5i4psmdawp7xsq

An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design [article]

Yi Zhang
2021 arXiv   pre-print
Concretely, 13 of them leverage GNN for molecular property prediction and 29 of them use RL and/or deep generative models for molecule generation and optimization.  ...  Moreover, 60 additional applications of AI in molecule generation and optimization are briefly summarized in a table.  ...  Mol-CycleGAN (Maziarka et al., 2020). Mol-CycleGAN is a GAN-based model that optimizes molecules while keeping them similar to the starting molecules.  ... 
arXiv:2110.05478v1 fatcat:uq2lera3znhpziuig6n44uuaby

Learning to Discover Medicines [article]

Tri Minh Nguyen, Thin Nguyen, Truyen Tran
2022 arXiv   pre-print
reasoning where we predict molecular properties and their binding, optimize existing compounds, generate de novo molecules, and plan the synthesis of target molecules; and (c) knowledge-based reasoning  ...  We organize the vast and rapidly growing literature of AI for drug discovery into three relatively stable sub-areas: (a) representation learning over molecular sequences and geometric graphs; (b) data-driven  ...  Molecular optimization and generation The second question is what kind of molecules interact with the given target.  ... 
arXiv:2202.07096v1 fatcat:u77zls6hezffbkmm3zy2rhnueu

Deep Graph Generators: A Survey

Faezeh Faez, Yassaman Ommi, Mahdieh Soleymani Baghshah, Hamid R. Rabiee
2021 IEEE Access  
novel molecular structures to modeling social networks.  ...  Deep generative models have achieved great success in areas such as image, speech, and natural language processing in the past few years.  ...  Mol-CycleGAN [69] establishes structural similarity between the source and target molecular graphs by adopting a CycleGAN-based [126] approach.  ... 
doi:10.1109/access.2021.3098417 fatcat:6xzg5cs75zhovdbpjkignfr3xu

AI for Chemical Space Gap Filling and Novel Compound Generation [article]

Monee Y. McGrady, Sean M. Colby, Jamie R Nuñez, Ryan S. Renslow, Thomas O. Metz
2022 arXiv   pre-print
With DarkChem's design, distance in this latent space is often associated with compound similarity, making sparse regions interesting targets for compound generation due to the possibility of generating  ...  We have used one such tool, a deep-learning software called DarkChem, which learns a representation of the molecular structure of compounds by compressing them into a latent space.  ...  examples of such tools are Mol-CycleGAN, 24 a generative model that creates compounds with a similar structure to the input compound but with optimized property values for increased efficacy against target  ... 
arXiv:2201.12398v1 fatcat:acujjbrytnhffm6aijsmvllg5y

Generative chemistry: drug discovery with deep learning generative models [article]

Yuemin Bian, Xiang-Qun Xie
2020 arXiv   pre-print
Commonly used chemical databases, molecular representations, and tools in cheminformatics and machine learning are covered as the infrastructure for the generative chemistry.  ...  From the generation of original texts, images, and videos, to the scratching of novel molecular structures, the incredible creativity of deep learning generative models surprised us about the height machine  ...  Łukasz Maziarka et al. introduced Mol-CycleGAN for derivatives design and compound optimization 115 .  ... 
arXiv:2008.09000v1 fatcat:ivznoc4bsbfoderwr2ted76fiq

Practical Massively Parallel Monte-Carlo Tree Search Applied to Molecular Design [article]

Xiufeng Yang and Tanuj Kr Aasawat and Kazuki Yoshizoe
2021 arXiv   pre-print
However, while massively parallel computing is often used for training models, it is rarely used for searching solutions for combinatorial optimization problems.  ...  In this paper, we propose a novel massively parallel Monte-Carlo Tree Search (MP-MCTS) algorithm that works efficiently for 1,000 worker scale, and apply it to molecular design.  ...  use deep generative models (to generate candidate molecules), followed by optimization algorithms to focus on the promising candidates having desired molecular property (mainly Bayesian Optimization (  ... 
arXiv:2006.10504v3 fatcat:xbo2c4rcsnghlaciwg3m5jf6my

A Binded VAE for Inorganic Material Generation [article]

Fouad Oubari, Antoine de Mathelin, Rodrigue Décatoire, Mathilde Mougeot
2021 arXiv   pre-print
We show on a real issue of rubber compound design that the proposed approach outperforms the standard generative models which opens new perspectives for material design optimization.  ...  Furthermore, traditional generative model validation processes as visual verification, FID and Inception scores are tailored for images and cannot then be used as such in this context.  ...  Mol-cyclegan: a generative model for molecular optimization. Journal of Cheminformatics, 12(1):1–18, 2020. [16] Reuben R Shamir, Yuval Duchin, Jinyoung Kim, Guillermo Sapiro, and Noam Harel.  ... 
arXiv:2112.09570v1 fatcat:gqussf4g4jbz3c4xtqvimny23q

Theory of Quantum Generative Learning Models with Maximum Mean Discrepancy [article]

Yuxuan Du, Zhuozhuo Tu, Bujiao Wu, Xiao Yuan, Dacheng Tao
2022 arXiv   pre-print
Our work opens the avenue for quantitatively understanding the power of quantum generative learning models.  ...  The intrinsic probabilistic nature of quantum mechanics invokes endeavors of designing quantum generative learning models (QGLMs) with computational advantages over classical ones.  ...  Mol-cyclegan: a generative model for molecular opti-mization.  ... 
arXiv:2205.04730v1 fatcat:rogjqhbvljfu3lsdnxyh7jfowu

Concepts of Artificial Intelligence for Computer-Assisted Drug Discovery

Xin Yang, Yifei Wang, Ryan Byrne, Gisbert Schneider, Shengyong Yang
2019
Artificial intelligence (AI), and, in particular, deep learning as a subcategory of AI, provides opportunities for the discovery and development of innovative drugs.  ...  Finally, several challenges and limitations of the current methods are summarized, with a view to potential future directions for AI-assisted drug discovery and design. CONTENTS  ...  Mol-CycleGAN is a CycleGAN-based generative model 515 that operates in the latent space trained by JT-VAE, 516 a VAE trained on junction tree molecular representations rather than SMILES, with the  ... 
doi:10.3929/ethz-b-000367388 fatcat:5lno7hywmva6pg3nlufgln23se