MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem

Jonghwan Choi; Sangmin Seo; Jinuk Park; Sanghyun Park

doi:10.1109/BIBM52615.2021.9669668

MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem

Jonghwan Choi, Sangmin Seo, Jinuk Park, Sanghyun Park

Department of Computer Science

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

Deep generative models for molecular generation have accelerated the development of de novo drug design by introducing how to generate novel molecular structures expressed in simplified molecular-input line-entry system (SMILES) or molecular graph formats. Numerous drug design studies have proposed combinations of variational autoencoder (VAE) and autoregressive generators such as recurrent neural networks (RNNs) to generate SMILES strings. However, RNN-VAE has one notorious issue, called posterior collapse, in which different latent vectors produce indistinguishable molecular distributions. In this study, we proposed a Gumbel-Softmax-based generative model, MolBit, and a genetic algorithm-based molecular property optimization method. We confirmed that the proposed model avoided the posterior collapse problem and outperformed the existing drug design models with SMILES.

Original language	English
Title of host publication	Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021
Editors	Yufei Huang, Lukasz Kurgan, Feng Luo, Xiaohua Tony Hu, Yidong Chen, Edward Dougherty, Andrzej Kloczkowski, Yaohang Li
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	364-367
Number of pages	4
ISBN (Electronic)	9781665401265
DOIs	https://doi.org/10.1109/BIBM52615.2021.9669668
Publication status	Published - 2021
Event	2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021 - Virtual, Online, United States Duration: 2021 Dec 9 → 2021 Dec 12

Publication series

Name	Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

Conference

Conference	2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021
Country/Territory	United States
City	Virtual, Online
Period	21/12/9 → 21/12/12

Bibliographical note

Funding Information:
This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT) (IITP-2017-0-00477, (SW starlab) Research and development of the high performance in-memory distributed DBMS based on flash memory storage in IoT environment). 978-1-6654-0126-5/21/$31.00 ©2021 IEEE

Publisher Copyright:
© 2021 IEEE.

All Science Journal Classification (ASJC) codes

Artificial Intelligence
Computer Science Applications
Biomedical Engineering
Health Informatics
Information Systems and Management

Access to Document

10.1109/BIBM52615.2021.9669668

Cite this

Choi, J., Seo, S., Park, J., & Park, S. (2021). MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem. In Y. Huang, L. Kurgan, F. Luo, X. T. Hu, Y. Chen, E. Dougherty, A. Kloczkowski, & Y. Li (Eds.), Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021 (pp. 364-367). (Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/BIBM52615.2021.9669668

Choi, Jonghwan ; Seo, Sangmin ; Park, Jinuk ; Park, Sanghyun. / MolBit : De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem. Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021. editor / Yufei Huang ; Lukasz Kurgan ; Feng Luo ; Xiaohua Tony Hu ; Yidong Chen ; Edward Dougherty ; Andrzej Kloczkowski ; Yaohang Li. Institute of Electrical and Electronics Engineers Inc., 2021. pp. 364-367 (Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021).

@inproceedings{e177fc3b86c94ba387ad4bcd727a2dd8,

title = "MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem",

abstract = "Deep generative models for molecular generation have accelerated the development of de novo drug design by introducing how to generate novel molecular structures expressed in simplified molecular-input line-entry system (SMILES) or molecular graph formats. Numerous drug design studies have proposed combinations of variational autoencoder (VAE) and autoregressive generators such as recurrent neural networks (RNNs) to generate SMILES strings. However, RNN-VAE has one notorious issue, called posterior collapse, in which different latent vectors produce indistinguishable molecular distributions. In this study, we proposed a Gumbel-Softmax-based generative model, MolBit, and a genetic algorithm-based molecular property optimization method. We confirmed that the proposed model avoided the posterior collapse problem and outperformed the existing drug design models with SMILES. ",

author = "Jonghwan Choi and Sangmin Seo and Jinuk Park and Sanghyun Park",

note = "Funding Information: This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT) (IITP-2017-0-00477, (SW starlab) Research and development of the high performance in-memory distributed DBMS based on flash memory storage in IoT environment). 978-1-6654-0126-5/21/$31.00 {\textcopyright}2021 IEEE Publisher Copyright: {\textcopyright} 2021 IEEE.; 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021 ; Conference date: 09-12-2021 Through 12-12-2021",

year = "2021",

doi = "10.1109/BIBM52615.2021.9669668",

language = "English",

series = "Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "364--367",

editor = "Yufei Huang and Lukasz Kurgan and Feng Luo and Hu, {Xiaohua Tony} and Yidong Chen and Edward Dougherty and Andrzej Kloczkowski and Yaohang Li",

booktitle = "Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021",

address = "United States",

}

Choi, J, Seo, S, Park, J & Park, S 2021, MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem. in Y Huang, L Kurgan, F Luo, XT Hu, Y Chen, E Dougherty, A Kloczkowski & Y Li (eds), Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021. Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021, Institute of Electrical and Electronics Engineers Inc., pp. 364-367, 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021, Virtual, Online, United States, 21/12/9. https://doi.org/10.1109/BIBM52615.2021.9669668

MolBit : De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem. / Choi, Jonghwan; Seo, Sangmin; Park, Jinuk; Park, Sanghyun.

Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021. ed. / Yufei Huang; Lukasz Kurgan; Feng Luo; Xiaohua Tony Hu; Yidong Chen; Edward Dougherty; Andrzej Kloczkowski; Yaohang Li. Institute of Electrical and Electronics Engineers Inc., 2021. p. 364-367 (Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

TY - GEN

T1 - MolBit

T2 - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

AU - Choi, Jonghwan

AU - Seo, Sangmin

AU - Park, Jinuk

AU - Park, Sanghyun

N1 - Funding Information: This work was supported by Institute of Information & communications Technology Planning & Evaluation (IITP) grant funded by the Korea government(MSIT) (IITP-2017-0-00477, (SW starlab) Research and development of the high performance in-memory distributed DBMS based on flash memory storage in IoT environment). 978-1-6654-0126-5/21/$31.00 ©2021 IEEE Publisher Copyright: © 2021 IEEE.

PY - 2021

Y1 - 2021

N2 - Deep generative models for molecular generation have accelerated the development of de novo drug design by introducing how to generate novel molecular structures expressed in simplified molecular-input line-entry system (SMILES) or molecular graph formats. Numerous drug design studies have proposed combinations of variational autoencoder (VAE) and autoregressive generators such as recurrent neural networks (RNNs) to generate SMILES strings. However, RNN-VAE has one notorious issue, called posterior collapse, in which different latent vectors produce indistinguishable molecular distributions. In this study, we proposed a Gumbel-Softmax-based generative model, MolBit, and a genetic algorithm-based molecular property optimization method. We confirmed that the proposed model avoided the posterior collapse problem and outperformed the existing drug design models with SMILES.

AB - Deep generative models for molecular generation have accelerated the development of de novo drug design by introducing how to generate novel molecular structures expressed in simplified molecular-input line-entry system (SMILES) or molecular graph formats. Numerous drug design studies have proposed combinations of variational autoencoder (VAE) and autoregressive generators such as recurrent neural networks (RNNs) to generate SMILES strings. However, RNN-VAE has one notorious issue, called posterior collapse, in which different latent vectors produce indistinguishable molecular distributions. In this study, we proposed a Gumbel-Softmax-based generative model, MolBit, and a genetic algorithm-based molecular property optimization method. We confirmed that the proposed model avoided the posterior collapse problem and outperformed the existing drug design models with SMILES.

UR - http://www.scopus.com/inward/record.url?scp=85125188506&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85125188506&partnerID=8YFLogxK

U2 - 10.1109/BIBM52615.2021.9669668

DO - 10.1109/BIBM52615.2021.9669668

M3 - Conference contribution

AN - SCOPUS:85125188506

T3 - Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

SP - 364

EP - 367

BT - Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021

A2 - Huang, Yufei

A2 - Kurgan, Lukasz

A2 - Luo, Feng

A2 - Hu, Xiaohua Tony

A2 - Chen, Yidong

A2 - Dougherty, Edward

A2 - Kloczkowski, Andrzej

A2 - Li, Yaohang

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 9 December 2021 through 12 December 2021

ER -

Choi J, Seo S, Park J, Park S. MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem. In Huang Y, Kurgan L, Luo F, Hu XT, Chen Y, Dougherty E, Kloczkowski A, Li Y, editors, Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021. Institute of Electrical and Electronics Engineers Inc. 2021. p. 364-367. (Proceedings - 2021 IEEE International Conference on Bioinformatics and Biomedicine, BIBM 2021). https://doi.org/10.1109/BIBM52615.2021.9669668

MolBit: De novo Drug Design via Binary Representations of SMILES for avoiding the Posterior Collapse Problem

Abstract

Publication series

Conference

Bibliographical note

All Science Journal Classification (ASJC) codes

Access to Document

Other files and links

Fingerprint

Cite this