Reinforcement learning-inspired molecular generation with latent space diffusion and genetic algorithm optimization under affinity and similarity constraints

•A novel RL-inspired framework combines VAE and latent-space diffusion for molecule generation.•Affinity and similarity constraints guide generation toward biologically active candidates.•A genetic algorithm with active learning enables iterative, reward-driven optimization. In deep learning-driven...

Full description

Saved in:
Bibliographic Details
Published in:Chemical engineering science Vol. 320; p. 122575
Main Authors: Bai, Can, Wu, Zijian, Han, Xianjun, Huang, Renbao
Format: Journal Article
Language:English
Published: Elsevier Ltd 15.01.2026
Subjects:
ISSN:0009-2509
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Abstract •A novel RL-inspired framework combines VAE and latent-space diffusion for molecule generation.•Affinity and similarity constraints guide generation toward biologically active candidates.•A genetic algorithm with active learning enables iterative, reward-driven optimization. In deep learning-driven molecular generation, achieving both diversity and effectiveness remains a major challenge. Inspired by the principles of reinforcement learning, in this study, a molecular generation system that integrates an encoding–diffusion–decoding mechanism with iterative, feedback-driven optimization strategies to balance these two aspects is proposed. The framework first maps molecular structures into a low-dimensional latent space, where a diffusion model explores the distribution of molecular characteristics. Sampling from a Gaussian distribution and performing reverse decoding ensure diversity in the molecular generation process. To ensure the practical applicability of the generated molecules, we incorporate a target-drug affinity prediction model and molecular similarity constraints into the pipeline to effectively filter candidates that are both novel and biologically relevant. Furthermore, a molecular genetic algorithm that mimics the exploration and exploitation trade-off that is fundamental to reinforcement learning is employed to perform random crossover and mutation on selected molecules, thereby generating potentially superior candidates. Guided by an active learning strategy, these candidates are iteratively evaluated and integrated into the training set, thus forming a continuous feedback loop that refines the generation model over time. This reinforcement learning-inspired framework not only increases the quality and efficiency of molecular generation but also reduces dependency on large, high-quality datasets. The experimental results confirm the ability of the method to generate effective and diverse compounds that target specific receptors.
AbstractList •A novel RL-inspired framework combines VAE and latent-space diffusion for molecule generation.•Affinity and similarity constraints guide generation toward biologically active candidates.•A genetic algorithm with active learning enables iterative, reward-driven optimization. In deep learning-driven molecular generation, achieving both diversity and effectiveness remains a major challenge. Inspired by the principles of reinforcement learning, in this study, a molecular generation system that integrates an encoding–diffusion–decoding mechanism with iterative, feedback-driven optimization strategies to balance these two aspects is proposed. The framework first maps molecular structures into a low-dimensional latent space, where a diffusion model explores the distribution of molecular characteristics. Sampling from a Gaussian distribution and performing reverse decoding ensure diversity in the molecular generation process. To ensure the practical applicability of the generated molecules, we incorporate a target-drug affinity prediction model and molecular similarity constraints into the pipeline to effectively filter candidates that are both novel and biologically relevant. Furthermore, a molecular genetic algorithm that mimics the exploration and exploitation trade-off that is fundamental to reinforcement learning is employed to perform random crossover and mutation on selected molecules, thereby generating potentially superior candidates. Guided by an active learning strategy, these candidates are iteratively evaluated and integrated into the training set, thus forming a continuous feedback loop that refines the generation model over time. This reinforcement learning-inspired framework not only increases the quality and efficiency of molecular generation but also reduces dependency on large, high-quality datasets. The experimental results confirm the ability of the method to generate effective and diverse compounds that target specific receptors.
ArticleNumber 122575
Author Bai, Can
Wu, Zijian
Huang, Renbao
Han, Xianjun
Author_xml – sequence: 1
  givenname: Can
  surname: Bai
  fullname: Bai, Can
  organization: College of Acupuncture and Tuina, Anhui University of Chinese Medicine, Hefei, China
– sequence: 2
  givenname: Zijian
  surname: Wu
  fullname: Wu, Zijian
  organization: College of Acupuncture and Tuina, Anhui University of Chinese Medicine, Hefei, China
– sequence: 3
  givenname: Xianjun
  orcidid: 0000-0001-7674-1428
  surname: Han
  fullname: Han, Xianjun
  email: hxj@ahu.edu.cn
  organization: School of Computer Science and Technology, Anhui University, Hefei, China
– sequence: 4
  givenname: Renbao
  surname: Huang
  fullname: Huang, Renbao
  organization: First Clinical Medical College, Anhui University of Chinese Medicine, Hefei, China
BookMark eNp9kEtOAzEMQLMoEm3hAOxygSlJOh9GrFDFT6qEhGAdZRKnuJpJqiQFlbtwVzIta1a2ZT_bejMycd4BIVecLTjj9fV2oSEuBBPVggtRNdWETBljbSEq1p6TWYzbXDYNZ1Py8wrorA8aBnCJ9qCCQ7cp0MUdBjB08D3ofa8C3YCDoBJ6R78wfdBepRGJO6WBGrR2H8eecuY4mlBT1W98yLMD9buEA36f8L0zEKiyFh2mw5GIuZuPjKX2Lqag0KV4Qc6s6iNc_sU5eX-4f1s9FeuXx-fV3brQouKpWGpTa1ubuu6s4Y1pc9qBUNy2SnVQqmWtm9YK1hh-ozvRibLhpa5NaZZ1qarlnPDTXh18jAGs3AUcVDhIzuToVG5ldipHp_LkNDO3JwbyY58IQUaN4DSY7E0naTz-Q_8CcZKJ2Q
Cites_doi 10.1093/nar/gkw1074
10.1038/s42256-023-00712-7
10.1109/TNNLS.2022.3147790
10.1038/s42256-021-00301-6
10.1002/advs.202417726
10.1016/j.compbiomed.2023.107372
10.1517/17460441.2015.1032936
10.1016/j.artmed.2024.102820
10.1109/TII.2023.3321332
10.1093/bib/bbad419
10.1093/bioinformatics/btz339
10.1007/s44196-024-00531-7
10.1021/ct3010485
10.1109/CVPR52688.2022.01042
10.1016/j.neunet.2024.106596
10.1007/s42979-021-00702-9
10.1093/nar/gkae300
10.1021/acs.jcim.1c00203
10.1038/sdata.2014.22
10.1186/1758-2946-4-24
10.1021/acs.jcim.2c00997
10.3389/frai.2024.1374148
10.26599/BDMA.2023.9020009
10.1016/j.compbiomed.2024.108702
10.1109/TAI.2023.3287947
10.1038/s42256-022-00557-6
10.1126/sciadv.aap7885
10.1021/acscentsci.7b00512
10.1186/s13321-017-0235-x
10.1186/s13321-021-00566-4
10.1021/ar000033j
10.1016/j.compbiomed.2024.108073
10.1021/acs.jcim.3c01964
10.1002/jcc.20035
10.1007/s10462-024-10714-5
10.3390/a17010024
10.1002/jcc.10128
10.1016/j.asoc.2024.111452
10.1016/j.simpa.2024.100656
10.1038/s43588-024-00737-x
10.1007/s10489-023-05217-9
10.1109/ACCESS.2024.3397775
10.1002/minf.202100045
10.1109/ACCESS.2024.3391249
10.1007/s10462-023-10687-x
10.1021/acscentsci.7b00572
ContentType Journal Article
Copyright 2025 Elsevier Ltd
Copyright_xml – notice: 2025 Elsevier Ltd
DBID AAYXX
CITATION
DOI 10.1016/j.ces.2025.122575
DatabaseName CrossRef
DatabaseTitle CrossRef
DatabaseTitleList
DeliveryMethod fulltext_linktorsrc
Discipline Engineering
ExternalDocumentID 10_1016_j_ces_2025_122575
S000925092501396X
GroupedDBID --K
--M
-~X
.~1
0R~
1B1
1~.
1~5
29B
4.4
457
4G.
5GY
5VS
6J9
7-5
71M
8P~
9JN
AABNK
AAEDT
AAEDW
AAIKJ
AAKOC
AALRI
AAOAW
AAQFI
AATTM
AAXKI
AAXUO
AAYWO
ABFNM
ABFRF
ABJNI
ABMAC
ABNUV
ACBEA
ACDAQ
ACGFO
ACGFS
ACLOT
ACNCT
ACRLP
ACVFH
ADBBV
ADCNI
ADEWK
ADEZE
AEBSH
AEFWE
AEIPS
AEKER
AENEX
AEUPX
AFJKZ
AFPUW
AFTJW
AGHFR
AGUBO
AGYEJ
AHHHB
AHPOS
AIEXJ
AIGII
AIIUN
AIKHN
AITUG
AKBMS
AKRWK
AKURH
AKYEP
ALMA_UNASSIGNED_HOLDINGS
AMRAJ
ANKPU
APXCP
AXJTR
BKOJK
BLXMC
CS3
DU5
EBS
EFJIC
EFKBS
EFLBG
ENUVR
EO8
EO9
EP2
EP3
F5P
FDB
FIRID
FNPLU
FYGXN
G-Q
GBLVA
HLY
IHE
J1W
KOM
LX7
M41
MO0
N9A
O-L
O9-
OAUVE
OZT
P-8
P-9
P2P
PC.
Q38
RNS
ROL
RPZ
SCE
SDF
SDG
SDP
SES
SEW
SPC
SPCBC
SSG
SSZ
T5K
XPP
ZMT
~02
~G-
~HD
9DU
AAQXK
AAYXX
ABDPE
ABWVN
ABXDB
ACRPL
ADMUD
ADNMO
AGQPQ
AI.
AIDUJ
ASPBG
AVWKF
AZFZN
BBWZM
CITATION
EJD
FEDTE
FGOYB
HVGLF
HZ~
NDZJH
R2-
SC5
T9H
VH1
WUQ
Y6R
ZY4
ID FETCH-LOGICAL-c251t-3cd6cf6d66bfd17d96d6be2a1f9aabe4a36c79f207d18cb2b24714c6d4d364a53
ISICitedReferencesCount 0
ISICitedReferencesURI http://www.webofscience.com/api/gateway?GWVersion=2&SrcApp=Summon&SrcAuth=ProQuest&DestLinkType=CitingArticles&DestApp=WOS_CPL&KeyUT=001578765800005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
ISSN 0009-2509
IngestDate Sat Nov 29 06:48:07 EST 2025
Wed Dec 10 14:31:34 EST 2025
IsPeerReviewed true
IsScholarly true
Keywords Genetic algorithm optimization
Affinity constraints
Molecular generation
Latent space generation
Language English
LinkModel OpenURL
MergedId FETCHMERGED-LOGICAL-c251t-3cd6cf6d66bfd17d96d6be2a1f9aabe4a36c79f207d18cb2b24714c6d4d364a53
ORCID 0000-0001-7674-1428
ParticipantIDs crossref_primary_10_1016_j_ces_2025_122575
elsevier_sciencedirect_doi_10_1016_j_ces_2025_122575
PublicationCentury 2000
PublicationDate 2026-01-15
PublicationDateYYYYMMDD 2026-01-15
PublicationDate_xml – month: 01
  year: 2026
  text: 2026-01-15
  day: 15
PublicationDecade 2020
PublicationTitle Chemical engineering science
PublicationYear 2026
Publisher Elsevier Ltd
Publisher_xml – name: Elsevier Ltd
References Li, Wang, Sun, Zeng, Yuan, Gou, Wang, Guo, Pu (bib0036) 2022; 62
Huang, L., Zhang, H., Xu, T., Wong, K.-c., 2022. Mdm: molecular diffusion model for 3d molecule generation.
Li, Yao, Wei, Niu, Zeng, Li, Wang (bib0037) 2024; 35
Segler, Kogej, Tyrchan, Waller (bib0057) 2018; 4
Liu, Xu, Luo, Jiang (bib0041) 2024
Nguyen, Roe, Simmerling (bib0046) 2013; 9
Olivecrona, Blaschke, Engkvist, Chen (bib0047) 2017; 9
Fu, Du, Zhang (bib0017) 2024; 62
Xu, Lei, Ma, Pan (bib0065) 2024; 7
Abbasi, Carvalho, Ribeiro, Arrais (bib0001) 2024; 150
Molnár, Tamás (bib0045) 2024; 57
Scutari (bib0056) 2024; 17
Chinnareddy, Grandhi, Narayan (bib0010) 2021
Assaf, Liu, Lin, Erkens (bib0003) 2024; 20
Genetic algorithm-based receptor ligand: A genetic algorithm-guided generative model to boost the novelty and drug-likeness of molecules in a sampling chemical space. 2024. J. Chem. Inf. Model. 64 (4), 1213–1228.
Zeng, Xiang, Yu, Wang, Li, Nussinov, Cheng (bib0067) 2022; 4
Quan, Yuan, Luo, Song, Zhou, Wang (bib0050) 2024; 20
Axelrod, S., Gómez-Bombarelli, R., 2020. Geom: energy-annotated molecular conformations for property prediction and molecular generation.
Lee, Kahng, Kim (bib0035) 2021; 40
Zhang, Xu, Xiao (bib0068) 2024; 17
Prasad, V., Chen, Z., Vilanova, A., Pfister, H., Pezzotti, N., Strobelt, H., 2023. Unraveling the temporal dynamics of the unet in diffusion models. 2312.14965.
Eberhardt, Santos-Martins, Tillack, Forli (bib0014) 2021; 61
Bugnon, Röhrig, Goullieux, Perez, Daina, Michielin, Zoete (bib0008) 2024; 52
.
Trabelsi, Chaabane, Ben-Hur (bib0059) 2019; 35
Wang, Zhu (bib0062) 2024; 179
Hoogeboom, E., Satorras, V. G., Vignac, C., Welling, M., 2022a. Equivariant diffusion for molecule generation in 3d.
Shen, Zeng, Zhu, Wang, Chen (bib0058) 2021; 3
Zhang, Zitnik, Liu (bib0070) 2024; 37
Jin, W., Barzilay, R., Jaakkola, T., 2018. Junction tree variational autoencoder for molecular graph generation.
Luo, Yan, Ji (bib0044) 2021
Lu, Xiao (bib0043) 2024; 54
Ramakrishnan, Dral, Rupp, Von Lilienfeld (bib0051) 2014; 1
Gholap, Uddin, Faiyazuddin, Omri, Gowri, Khalid (bib0023) 2024; 178
Ganea, Pattanaik, Coley, Barzilay, Jensen, Green, Jaakkola (bib0018) 2021; 34
Lin, Zhang, Duan, Ou-Yang, Zhao (bib0039) 2023
Li, Gao, Gao (bib0038) 2024
Dong, Wu, Xu, Ouyang (bib0013) 2023; 25
Asperti, Evangelista, Piccolomini (bib0002) 2021; 2
Gebauer, N. W. A., Gastegger, M., Schütt, K. T., 2019. Symmetry-adapted generation of 3D point sets for the targeted discovery of molecules. Adv. Neural Inf. Process. Syst. 32, 10757–10769.
Schütt, Kindermans, Sauceda Felix, Chmiela, Tkatchenko, Müller (bib0055) 2017; 30
Yang, Gu, Liu, Gong, Lu, Qiu, Yao, Liu (bib0066) 2025; 12
Pushkaran, Arabi (bib0049) 2024; 57
Guo, Pu, Jiao, Peng, Wang, Yang (bib0026) 2024; 155
Gómez-Bombarelli, Duvenaud, Hernández-Lobato, Aguilera-Iparraguirre, Hirzel, Adams, Aspuru-Guzik (bib0024) 2016; 4
Xu, M., Yu, L., Song, Y., Shi, C., Ermon, S., Tang, J., 2022. Geodiff: a geometric diffusion model for molecular conformation generation.
Bengesi, El-Sayed, Sarker, Houkpati, Irungu, Oladunni (bib0005) 2024; 12
Rombach, R., Blattmann, A., Lorenz, D., Esser, P., Ommer, B., 2022. High-resolution image synthesis with latent diffusion models. 2112.10752.
Kollman, Massova, Reyes, Kuhn, Huo, Chong, Lee, Lee, Duan, Wang (bib0034) 2000; 33
Satorras, Hoogeboom, Fuchs, Posner, Welling (bib0053) 2021
van de Meent, Bronson, Wood, Gonzalez, Wiggins (bib0060) 2013
Fang, Zhang, Du, He (bib0016) 2023; 164
Gaulton, Hersey, Nowotka, Bento, Chambers, Mendez, Mutowo, Atkinson, Bellis, Cibrián-Uhalte, Davies, Dedman, Karlsson, Magariños, Overington, Papadatos, Smit, Leach (bib0019) 2016; 45
Genheden, Ryde (bib0022) 2015; 10
Deep reinforcement learning for de novo drug design, 2017. Sci. Adv. 4.
Hoogeboom, E., Satorras, V. G., Vignac, C., Welling, M., 2022b. Equivariant diffusion for molecule generation in 3D. Proceedings of the 39th International Conference on Machine Learning, 1628867–8887.
Wang, Wolf, Caldwell, Kollman, Case (bib0061) 2004; 25
Wang, W., Bao, J., Zhou, W., Chen, D., Chen, D., Yuan, L., Li, H., 2022. Sindiffusion: Learning a diffusion model from a single natural image. 2211.12445.
Chaturvedi, Dhiman, Vishwakarma (bib0009) 2024; 41
Zhang, Zhang, Jin, Zhang, Hu, Shen, Cao, Du, Kang, Deng (bib0069) 2023; 5
Ho, J., Jain, A., Abbeel, P., 2020. Denoising diffusion probabilistic models. 2006.11239.
Liu, Song, Na, Wang (bib0040) 2024; 7
Cui (bib0011) 2024; 12
Guimaraes, G. L., Sánchez-Lengeling, B., Farias, P. L. C., Aspuru-Guzik, A., 2017. Objective-reinforced generative adversarial networks (ORGAN) for sequence generation models.
Berenger, Tsuda (bib0006) 2021; 13
Jakalian, Jack, Bayly (bib0032) 2002; 23
Brefo-Mensah, Palmer (bib0007) 2012; 4
Hu, Malkin, Jain, Everett, Graikos, Bengio (bib0030) 2023
Ellis, Iqbal, Yoshimatsu (bib0015) 2024; 5
Liu, Ren, Tao, Ren (bib0042) 2024; 171
Schneuing, Harris, Du, Didi, Jamasb, Igashov, Du, Gomes, Blundell, Lio (bib0054) 2024; 4
Brefo-Mensah (10.1016/j.ces.2025.122575_bib0007) 2012; 4
Luo (10.1016/j.ces.2025.122575_bib0044) 2021
Liu (10.1016/j.ces.2025.122575_bib0040) 2024; 7
Li (10.1016/j.ces.2025.122575_bib0037) 2024; 35
Satorras (10.1016/j.ces.2025.122575_bib0053) 2021
Molnár (10.1016/j.ces.2025.122575_bib0045) 2024; 57
Cui (10.1016/j.ces.2025.122575_bib0011) 2024; 12
Lin (10.1016/j.ces.2025.122575_bib0039) 2023
10.1016/j.ces.2025.122575_bib0004
10.1016/j.ces.2025.122575_bib0048
Genheden (10.1016/j.ces.2025.122575_bib0022) 2015; 10
Pushkaran (10.1016/j.ces.2025.122575_bib0049) 2024; 57
Shen (10.1016/j.ces.2025.122575_bib0058) 2021; 3
Hu (10.1016/j.ces.2025.122575_bib0030) 2023
Gaulton (10.1016/j.ces.2025.122575_bib0019) 2016; 45
10.1016/j.ces.2025.122575_bib0052
10.1016/j.ces.2025.122575_bib0012
Segler (10.1016/j.ces.2025.122575_bib0057) 2018; 4
Ganea (10.1016/j.ces.2025.122575_bib0018) 2021; 34
Zhang (10.1016/j.ces.2025.122575_bib0069) 2023; 5
Wang (10.1016/j.ces.2025.122575_bib0062) 2024; 179
Ellis (10.1016/j.ces.2025.122575_bib0015) 2024; 5
Jakalian (10.1016/j.ces.2025.122575_bib0032) 2002; 23
Fu (10.1016/j.ces.2025.122575_bib0017) 2024; 62
Nguyen (10.1016/j.ces.2025.122575_bib0046) 2013; 9
Gómez-Bombarelli (10.1016/j.ces.2025.122575_bib0024) 2016; 4
Lee (10.1016/j.ces.2025.122575_bib0035) 2021; 40
Schütt (10.1016/j.ces.2025.122575_bib0055) 2017; 30
10.1016/j.ces.2025.122575_bib0063
Zhang (10.1016/j.ces.2025.122575_bib0068) 2024; 17
Chinnareddy (10.1016/j.ces.2025.122575_bib0010) 2021
van de Meent (10.1016/j.ces.2025.122575_bib0060) 2013
10.1016/j.ces.2025.122575_bib0020
10.1016/j.ces.2025.122575_bib0064
10.1016/j.ces.2025.122575_bib0021
Berenger (10.1016/j.ces.2025.122575_bib0006) 2021; 13
Ramakrishnan (10.1016/j.ces.2025.122575_bib0051) 2014; 1
Eberhardt (10.1016/j.ces.2025.122575_bib0014) 2021; 61
Xu (10.1016/j.ces.2025.122575_bib0065) 2024; 7
Fang (10.1016/j.ces.2025.122575_bib0016) 2023; 164
Yang (10.1016/j.ces.2025.122575_bib0066) 2025; 12
Chaturvedi (10.1016/j.ces.2025.122575_bib0009) 2024; 41
Kollman (10.1016/j.ces.2025.122575_bib0034) 2000; 33
Wang (10.1016/j.ces.2025.122575_bib0061) 2004; 25
Liu (10.1016/j.ces.2025.122575_bib0042) 2024; 171
Scutari (10.1016/j.ces.2025.122575_bib0056) 2024; 17
10.1016/j.ces.2025.122575_bib0027
Assaf (10.1016/j.ces.2025.122575_bib0003) 2024; 20
Bengesi (10.1016/j.ces.2025.122575_bib0005) 2024; 12
10.1016/j.ces.2025.122575_bib0025
Zhang (10.1016/j.ces.2025.122575_bib0070) 2024; 37
Gholap (10.1016/j.ces.2025.122575_bib0023) 2024; 178
10.1016/j.ces.2025.122575_bib0028
10.1016/j.ces.2025.122575_bib0029
Quan (10.1016/j.ces.2025.122575_bib0050) 2024; 20
10.1016/j.ces.2025.122575_bib0033
10.1016/j.ces.2025.122575_bib0031
Li (10.1016/j.ces.2025.122575_bib0036) 2022; 62
Lu (10.1016/j.ces.2025.122575_bib0043) 2024; 54
Olivecrona (10.1016/j.ces.2025.122575_bib0047) 2017; 9
Li (10.1016/j.ces.2025.122575_bib0038) 2024
Guo (10.1016/j.ces.2025.122575_bib0026) 2024; 155
Zeng (10.1016/j.ces.2025.122575_bib0067) 2022; 4
Schneuing (10.1016/j.ces.2025.122575_bib0054) 2024; 4
Asperti (10.1016/j.ces.2025.122575_bib0002) 2021; 2
Dong (10.1016/j.ces.2025.122575_bib0013) 2023; 25
Trabelsi (10.1016/j.ces.2025.122575_bib0059) 2019; 35
Bugnon (10.1016/j.ces.2025.122575_bib0008) 2024; 52
Liu (10.1016/j.ces.2025.122575_bib0041) 2024
Abbasi (10.1016/j.ces.2025.122575_bib0001) 2024; 150
References_xml – volume: 4
  start-page: 268
  year: 2016
  end-page: 276
  ident: bib0024
  article-title: Automatic chemical design using a data-driven continuous representation of molecules
  publication-title: ACS Cent. Sci.
– reference: Deep reinforcement learning for de novo drug design, 2017. Sci. Adv. 4.
– reference: Ho, J., Jain, A., Abbeel, P., 2020. Denoising diffusion probabilistic models. 2006.11239.
– volume: 20
  start-page: 4137
  year: 2024
  end-page: 4148
  ident: bib0050
  article-title: From regression to classification: fuzzy multikernel subspace learning for robust prediction and drug screening
  publication-title: IEEE Trans. Ind. Inf.
– volume: 4
  start-page: 120
  year: 2018
  end-page: 131
  ident: bib0057
  article-title: Generating focused molecule libraries for drug discovery with recurrent neural networks
  publication-title: ACS Cent. Sci.
– volume: 1
  year: 2014
  ident: bib0051
  article-title: Quantum chemistry structures and properties of 134 kilo molecules
  publication-title: Sci. Data
– reference: Rombach, R., Blattmann, A., Lorenz, D., Esser, P., Ommer, B., 2022. High-resolution image synthesis with latent diffusion models. 2112.10752.
– reference: Xu, M., Yu, L., Song, Y., Shi, C., Ermon, S., Tang, J., 2022. Geodiff: a geometric diffusion model for molecular conformation generation.
– volume: 5
  start-page: 1020
  year: 2023
  end-page: 1030
  ident: bib0069
  article-title: Resgen is a pocket-aware 3D molecular generation model based on parallel multiscale modelling
  publication-title: Nature Mach. Intell.
– volume: 155
  year: 2024
  ident: bib0026
  article-title: Online semi-supervised active learning ensemble classification for evolving imbalanced data streams
  publication-title: Appl. Soft Comput.
– volume: 150
  year: 2024
  ident: bib0001
  article-title: Predicting drug activity against cancer through genomic profiles and SMILES
  publication-title: Artif. Intell. Medicine
– volume: 37
  start-page: 38559
  year: 2024
  end-page: 38589
  ident: bib0070
  article-title: Generalized protein pocket generation with prior-informed flow matching
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 4
  start-page: 24
  year: 2012
  ident: bib0007
  article-title: Mol2chemfig, a tool for rendering chemical structures from molfile or SMILES format to LATEX code
  publication-title: J. Cheminform.
– start-page: 361
  year: 2013
  end-page: 369
  ident: bib0060
  article-title: Hierarchically-coupled hidden Markov models for learning kinetic rates from single-molecule data
  publication-title: Proceedings of the 30th International Conference on Machine Learning, ICML 2013, Atlanta, GA, USA, 16–21 June 2013
– volume: 171
  year: 2024
  ident: bib0042
  article-title: Git-mol: a multi-modal large language model for molecular science with graph, image, and text
  publication-title: Comput. Biol. Medicine
– start-page: 57
  year: 2021
  end-page: 60
  ident: bib0010
  article-title: Self-attention mechanism in GANs for molecule generation
  publication-title: 20th IEEE International Conference on Machine Learning and Applications, ICMLA 2021, Pasadena, CA, USA, December 13–16, 2021
– volume: 35
  start-page: i269
  year: 2019
  end-page: i277
  ident: bib0059
  article-title: Comprehensive evaluation of deep learning architectures for prediction of DNA/RNA sequence binding specificities
  publication-title: Bioinformatics
– reference: Gebauer, N. W. A., Gastegger, M., Schütt, K. T., 2019. Symmetry-adapted generation of 3D point sets for the targeted discovery of molecules. Adv. Neural Inf. Process. Syst. 32, 10757–10769.
– volume: 3
  start-page: 334
  year: 2021
  end-page: 343
  ident: bib0058
  article-title: Out-of-the-box deep learning prediction of pharmaceutical properties by broadly learned knowledge-based molecular representations
  publication-title: Nature Mach. Intell.
– volume: 2
  start-page: 301
  year: 2021
  ident: bib0002
  article-title: A survey on variational autoencoders from a green AI perspective
  publication-title: SN Comput. Sci.
– volume: 25
  start-page: 1157
  year: 2004
  end-page: 1174
  ident: bib0061
  article-title: Development and testing of a general amber force field
  publication-title: J. Comput. Chem.
– volume: 33
  start-page: 889
  year: 2000
  end-page: 897
  ident: bib0034
  article-title: Calculating structures and free energies of complex molecules: combining molecular mechanics and continuum models
  publication-title: Acc. Chem. Res.
– start-page: 514
  year: 2023
  end-page: 522
  ident: bib0039
  article-title: MoVAE: a variational autoencoder for molecular graph generation
  publication-title: Proceedings of the 2023 SIAM International Conference on Data Mining, SDM 2023, Minneapolis-St. Paul Twin Cities, MN, USA, April 27–29, 2023
– volume: 17
  start-page: 24
  year: 2024
  ident: bib0056
  article-title: Entropy and the kullback-leibler divergence for bayesian networks: computational complexity and efficient implementation
  publication-title: Algorithms
– reference: Hoogeboom, E., Satorras, V. G., Vignac, C., Welling, M., 2022b. Equivariant diffusion for molecule generation in 3D. Proceedings of the 39th International Conference on Machine Learning, 1628867–8887.
– reference: .
– volume: 17
  start-page: 125
  year: 2024
  ident: bib0068
  article-title: STBGRN: A traffic prediction model based on spatiotemporal bidirectional gated recurrent units and graph convolutional residual networks
  publication-title: Int. J. Comput. Intell. Syst.
– volume: 178
  year: 2024
  ident: bib0023
  article-title: Advances in artificial intelligence for drug delivery and development: a comprehensive review
  publication-title: Comput. Biol. Medicine
– volume: 34
  start-page: 13757
  year: 2021
  end-page: 13769
  ident: bib0018
  article-title: Geomol: torsional geometric generation of molecular 3d conformer ensembles
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 164
  year: 2023
  ident: bib0016
  article-title: ColdDTA: utilizing data augmentation and attention-based feature fusion for drug-target binding affinity prediction
  publication-title: Comput. Biol. Med.
– volume: 7
  start-page: 142
  year: 2024
  end-page: 155
  ident: bib0065
  article-title: Molecular generation and optimization of molecular properties using a transformer model
  publication-title: Big Data Min. Anal.
– volume: 12
  start-page: 57071
  year: 2024
  end-page: 57081
  ident: bib0011
  article-title: An attention-based improved u-net neural network model for semantic segmentation of moving objects
  publication-title: IEEE Access
– volume: 54
  start-page: 985
  year: 2024
  end-page: 1002
  ident: bib0043
  article-title: A novel belief tanimoto coefficient with its applications in multisource information fusion
  publication-title: Appl. Intell.
– start-page: 7192
  year: 2021
  end-page: 7203
  ident: bib0044
  article-title: Graphdf: a discrete flow model for molecular graph generation
  publication-title: International Conference on Machine Learning
– volume: 45
  start-page: D945
  year: 2016
  end-page: D954
  ident: bib0019
  article-title: The ChEMBL database in 2017
  publication-title: Nucleic Acids Res.
– volume: 61
  start-page: 3891
  year: 2021
  end-page: 3898
  ident: bib0014
  article-title: Autodock vina 1.2. 0: new docking methods, expanded force field, and python bindings
  publication-title: J. Chem. Inf. Model.
– volume: 5
  start-page: 634
  year: 2024
  end-page: 646
  ident: bib0015
  article-title: Deep q-learning-based molecular graph generation for chemical structure prediction from infrared spectra
  publication-title: IEEE Trans. Artif. Intell.
– volume: 52
  start-page: W324
  year: 2024
  end-page: W332
  ident: bib0008
  article-title: Swissdock 2024: major enhancements for small-molecule docking with attracting cavities and autodock vina
  publication-title: Nucleic Acids Res.
– reference: Huang, L., Zhang, H., Xu, T., Wong, K.-c., 2022. Mdm: molecular diffusion model for 3d molecule generation.
– volume: 23
  start-page: 1623
  year: 2002
  end-page: 1641
  ident: bib0032
  article-title: Fast, efficient generation of high-quality atomic charges. AM1-BCC model: II. parameterization and validation
  publication-title: J. Comput. Chem.
– volume: 4
  start-page: 1004
  year: 2022
  end-page: 1016
  ident: bib0067
  article-title: Accurate prediction of molecular properties and drug targets using a self-supervised image representation learning framework
  publication-title: Nat. Mac. Intell.
– volume: 40
  year: 2021
  ident: bib0035
  article-title: Generative adversarial networks for de novo molecular design
  publication-title: Mol. Inform.
– year: 2021
  ident: bib0053
  article-title: E(n) equivariant normalizing flows
  publication-title: Neural Information Processing Systems
– volume: 12
  start-page: 69812
  year: 2024
  end-page: 69837
  ident: bib0005
  article-title: Advancements in generative AI: a comprehensive review of gans, gpt, autoencoders, diffusion model, and transformers
  publication-title: IEEE Access
– volume: 62
  start-page: 1
  year: 2024
  end-page: 13
  ident: bib0017
  article-title: Do we need learnable classifiers? a hyperspectral image classification algorithm based on attention-enhanced resblock-in-resblock and ETF classifier
  publication-title: IEEE Trans. Geosci. Remote. Sens.
– start-page: 13528
  year: 2023
  end-page: 13549
  ident: bib0030
  article-title: GFlowNet-EM for learning compositional latent variable models
  publication-title: Proceedings of the 40th International Conference on Machine Learning
– volume: 179
  year: 2024
  ident: bib0062
  article-title: Multi-objective molecular generation via clustered pareto-based reinforcement learning
  publication-title: Neural Netw.
– volume: 7
  year: 2024
  ident: bib0040
  article-title: PED: A novel predictor-encoder-decoder model for alzheimer drug molecular generation
  publication-title: Frontiers Artif. Intell.
– volume: 9
  start-page: 2020
  year: 2013
  end-page: 2034
  ident: bib0046
  article-title: Improved generalized born solvent model parameters for protein simulations
  publication-title: J. Chem. Theory Comput.
– reference: Wang, W., Bao, J., Zhou, W., Chen, D., Chen, D., Yuan, L., Li, H., 2022. Sindiffusion: Learning a diffusion model from a single natural image. 2211.12445.
– volume: 57
  start-page: 42
  year: 2024
  ident: bib0045
  article-title: Variational autoencoders for 3D data processing
  publication-title: Artif. Intell. Rev.
– reference: Genetic algorithm-based receptor ligand: A genetic algorithm-guided generative model to boost the novelty and drug-likeness of molecules in a sampling chemical space. 2024. J. Chem. Inf. Model. 64 (4), 1213–1228.
– reference: Guimaraes, G. L., Sánchez-Lengeling, B., Farias, P. L. C., Aspuru-Guzik, A., 2017. Objective-reinforced generative adversarial networks (ORGAN) for sequence generation models.
– volume: 35
  start-page: 4852
  year: 2024
  end-page: 4861
  ident: bib0037
  article-title: Geometry-based molecular generation with deep constrained variational autoencoder
  publication-title: IEEE Trans. Neural Networks Learn. Syst.
– year: 2024
  ident: bib0041
  article-title: Graph diffusion transformers for multi-conditional molecular generation
  publication-title: The Thirty-Eighth Annual Conference on Neural Information Processing Systems
– volume: 41
  year: 2024
  ident: bib0009
  article-title: Fight detection with spatial and channel wise attention-based convLSTM model
  publication-title: Expert Syst. J. Knowl. Eng.
– reference: Hoogeboom, E., Satorras, V. G., Vignac, C., Welling, M., 2022a. Equivariant diffusion for molecule generation in 3d.
– reference: Prasad, V., Chen, Z., Vilanova, A., Pfister, H., Pezzotti, N., Strobelt, H., 2023. Unraveling the temporal dynamics of the unet in diffusion models. 2312.14965.
– reference: .
– volume: 10
  start-page: 449
  year: 2015
  end-page: 461
  ident: bib0022
  article-title: The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities
  publication-title: Expert Opin. Drug Discov.
– volume: 62
  start-page: 4873
  year: 2022
  end-page: 4887
  ident: bib0036
  article-title: Correlated RNN framework to quickly generate molecules with desired properties for energetic materials in the low data regime
  publication-title: J. Chem. Inf. Model.
– volume: 12
  year: 2025
  ident: bib0066
  article-title: DiffMC-gen: a dual denoising diffusion model for multi-conditional molecular generation
  publication-title: Adv. Sci.
– volume: 57
  start-page: 86
  year: 2024
  ident: bib0049
  article-title: From understanding diseases to drug design: can artificial intelligence bridge the gap?
  publication-title: Artif. Intell. Rev.
– volume: 9
  year: 2017
  ident: bib0047
  article-title: Molecular de-novo design through deep reinforcement learning
  publication-title: J. Cheminform.
– volume: 20
  year: 2024
  ident: bib0003
  article-title: PDB2DAT: automating LAMMPS data file generation from PDB molecular systems using python, rdkit, and pysimm
  publication-title: Softw. Impacts
– reference: Axelrod, S., Gómez-Bombarelli, R., 2020. Geom: energy-annotated molecular conformations for property prediction and molecular generation.
– volume: 13
  start-page: 1
  year: 2021
  end-page: 10
  ident: bib0006
  article-title: Molecular generation by fast assembly of (deep)SMILES fragments
  publication-title: J. Cheminform.
– volume: 25
  year: 2023
  ident: bib0013
  article-title: FormulationAI: a novel web-based platform for drug formulation design driven by artificial intelligence
  publication-title: Briefings Bioinform.
– year: 2024
  ident: bib0038
  article-title: Point Cloud Compression - Technologies and Standardization
– volume: 4
  start-page: 899
  year: 2024
  end-page: 909
  ident: bib0054
  article-title: Structure-based drug design with equivariant diffusion models
  publication-title: Nature Comput. Sci.
– reference: Jin, W., Barzilay, R., Jaakkola, T., 2018. Junction tree variational autoencoder for molecular graph generation.
– volume: 30
  start-page: 809
  year: 2017
  end-page: 919
  ident: bib0055
  article-title: SCHNet: a continuous-filter convolutional neural network for modeling quantum interactions
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 45
  start-page: D945
  issue: D1
  year: 2016
  ident: 10.1016/j.ces.2025.122575_bib0019
  article-title: The ChEMBL database in 2017
  publication-title: Nucleic Acids Res.
  doi: 10.1093/nar/gkw1074
– volume: 5
  start-page: 1020
  issue: 9
  year: 2023
  ident: 10.1016/j.ces.2025.122575_bib0069
  article-title: Resgen is a pocket-aware 3D molecular generation model based on parallel multiscale modelling
  publication-title: Nature Mach. Intell.
  doi: 10.1038/s42256-023-00712-7
– volume: 35
  start-page: 4852
  issue: 4
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0037
  article-title: Geometry-based molecular generation with deep constrained variational autoencoder
  publication-title: IEEE Trans. Neural Networks Learn. Syst.
  doi: 10.1109/TNNLS.2022.3147790
– volume: 3
  start-page: 334
  issue: 4
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0058
  article-title: Out-of-the-box deep learning prediction of pharmaceutical properties by broadly learned knowledge-based molecular representations
  publication-title: Nature Mach. Intell.
  doi: 10.1038/s42256-021-00301-6
– ident: 10.1016/j.ces.2025.122575_bib0028
– year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0053
  article-title: E(n) equivariant normalizing flows
– volume: 12
  issue: 22
  year: 2025
  ident: 10.1016/j.ces.2025.122575_bib0066
  article-title: DiffMC-gen: a dual denoising diffusion model for multi-conditional molecular generation
  publication-title: Adv. Sci.
  doi: 10.1002/advs.202417726
– volume: 164
  year: 2023
  ident: 10.1016/j.ces.2025.122575_bib0016
  article-title: ColdDTA: utilizing data augmentation and attention-based feature fusion for drug-target binding affinity prediction
  publication-title: Comput. Biol. Med.
  doi: 10.1016/j.compbiomed.2023.107372
– volume: 10
  start-page: 449
  issue: 5
  year: 2015
  ident: 10.1016/j.ces.2025.122575_bib0022
  article-title: The MM/PBSA and MM/GBSA methods to estimate ligand-binding affinities
  publication-title: Expert Opin. Drug Discov.
  doi: 10.1517/17460441.2015.1032936
– volume: 34
  start-page: 13757
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0018
  article-title: Geomol: torsional geometric generation of molecular 3d conformer ensembles
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 150
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0001
  article-title: Predicting drug activity against cancer through genomic profiles and SMILES
  publication-title: Artif. Intell. Medicine
  doi: 10.1016/j.artmed.2024.102820
– volume: 62
  start-page: 1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0017
  article-title: Do we need learnable classifiers? a hyperspectral image classification algorithm based on attention-enhanced resblock-in-resblock and ETF classifier
  publication-title: IEEE Trans. Geosci. Remote. Sens.
– ident: 10.1016/j.ces.2025.122575_bib0063
– volume: 20
  start-page: 4137
  issue: 3
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0050
  article-title: From regression to classification: fuzzy multikernel subspace learning for robust prediction and drug screening
  publication-title: IEEE Trans. Ind. Inf.
  doi: 10.1109/TII.2023.3321332
– volume: 25
  issue: 1
  year: 2023
  ident: 10.1016/j.ces.2025.122575_bib0013
  article-title: FormulationAI: a novel web-based platform for drug formulation design driven by artificial intelligence
  publication-title: Briefings Bioinform.
  doi: 10.1093/bib/bbad419
– year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0041
  article-title: Graph diffusion transformers for multi-conditional molecular generation
– volume: 35
  start-page: i269
  issue: 14
  year: 2019
  ident: 10.1016/j.ces.2025.122575_bib0059
  article-title: Comprehensive evaluation of deep learning architectures for prediction of DNA/RNA sequence binding specificities
  publication-title: Bioinformatics
  doi: 10.1093/bioinformatics/btz339
– volume: 17
  start-page: 125
  issue: 1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0068
  article-title: STBGRN: A traffic prediction model based on spatiotemporal bidirectional gated recurrent units and graph convolutional residual networks
  publication-title: Int. J. Comput. Intell. Syst.
  doi: 10.1007/s44196-024-00531-7
– start-page: 57
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0010
  article-title: Self-attention mechanism in GANs for molecule generation
– volume: 9
  start-page: 2020
  issue: 4
  year: 2013
  ident: 10.1016/j.ces.2025.122575_bib0046
  article-title: Improved generalized born solvent model parameters for protein simulations
  publication-title: J. Chem. Theory Comput.
  doi: 10.1021/ct3010485
– ident: 10.1016/j.ces.2025.122575_bib0052
  doi: 10.1109/CVPR52688.2022.01042
– volume: 179
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0062
  article-title: Multi-objective molecular generation via clustered pareto-based reinforcement learning
  publication-title: Neural Netw.
  doi: 10.1016/j.neunet.2024.106596
– volume: 2
  start-page: 301
  issue: 4
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0002
  article-title: A survey on variational autoencoders from a green AI perspective
  publication-title: SN Comput. Sci.
  doi: 10.1007/s42979-021-00702-9
– ident: 10.1016/j.ces.2025.122575_bib0029
– start-page: 13528
  year: 2023
  ident: 10.1016/j.ces.2025.122575_bib0030
  article-title: GFlowNet-EM for learning compositional latent variable models
– volume: 52
  start-page: W324
  issue: W1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0008
  article-title: Swissdock 2024: major enhancements for small-molecule docking with attracting cavities and autodock vina
  publication-title: Nucleic Acids Res.
  doi: 10.1093/nar/gkae300
– ident: 10.1016/j.ces.2025.122575_bib0025
– volume: 30
  start-page: 809
  year: 2017
  ident: 10.1016/j.ces.2025.122575_bib0055
  article-title: SCHNet: a continuous-filter convolutional neural network for modeling quantum interactions
  publication-title: Adv. Neural Inf. Process. Syst.
– ident: 10.1016/j.ces.2025.122575_bib0004
– ident: 10.1016/j.ces.2025.122575_bib0031
– ident: 10.1016/j.ces.2025.122575_bib0021
– volume: 61
  start-page: 3891
  issue: 8
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0014
  article-title: Autodock vina 1.2. 0: new docking methods, expanded force field, and python bindings
  publication-title: J. Chem. Inf. Model.
  doi: 10.1021/acs.jcim.1c00203
– volume: 1
  year: 2014
  ident: 10.1016/j.ces.2025.122575_bib0051
  article-title: Quantum chemistry structures and properties of 134 kilo molecules
  publication-title: Sci. Data
  doi: 10.1038/sdata.2014.22
– volume: 4
  start-page: 24
  year: 2012
  ident: 10.1016/j.ces.2025.122575_bib0007
  article-title: Mol2chemfig, a tool for rendering chemical structures from molfile or SMILES format to LATEX code
  publication-title: J. Cheminform.
  doi: 10.1186/1758-2946-4-24
– start-page: 361
  year: 2013
  ident: 10.1016/j.ces.2025.122575_bib0060
  article-title: Hierarchically-coupled hidden Markov models for learning kinetic rates from single-molecule data
– ident: 10.1016/j.ces.2025.122575_bib0064
– volume: 62
  start-page: 4873
  issue: 20
  year: 2022
  ident: 10.1016/j.ces.2025.122575_bib0036
  article-title: Correlated RNN framework to quickly generate molecules with desired properties for energetic materials in the low data regime
  publication-title: J. Chem. Inf. Model.
  doi: 10.1021/acs.jcim.2c00997
– volume: 7
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0040
  article-title: PED: A novel predictor-encoder-decoder model for alzheimer drug molecular generation
  publication-title: Frontiers Artif. Intell.
  doi: 10.3389/frai.2024.1374148
– volume: 7
  start-page: 142
  issue: 1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0065
  article-title: Molecular generation and optimization of molecular properties using a transformer model
  publication-title: Big Data Min. Anal.
  doi: 10.26599/BDMA.2023.9020009
– volume: 178
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0023
  article-title: Advances in artificial intelligence for drug delivery and development: a comprehensive review
  publication-title: Comput. Biol. Medicine
  doi: 10.1016/j.compbiomed.2024.108702
– volume: 5
  start-page: 634
  issue: 2
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0015
  article-title: Deep q-learning-based molecular graph generation for chemical structure prediction from infrared spectra
  publication-title: IEEE Trans. Artif. Intell.
  doi: 10.1109/TAI.2023.3287947
– volume: 4
  start-page: 1004
  issue: 11
  year: 2022
  ident: 10.1016/j.ces.2025.122575_bib0067
  article-title: Accurate prediction of molecular properties and drug targets using a self-supervised image representation learning framework
  publication-title: Nat. Mac. Intell.
  doi: 10.1038/s42256-022-00557-6
– volume: 41
  issue: 1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0009
  article-title: Fight detection with spatial and channel wise attention-based convLSTM model
  publication-title: Expert Syst. J. Knowl. Eng.
– ident: 10.1016/j.ces.2025.122575_bib0012
  doi: 10.1126/sciadv.aap7885
– volume: 4
  start-page: 120
  issue: 1
  year: 2018
  ident: 10.1016/j.ces.2025.122575_bib0057
  article-title: Generating focused molecule libraries for drug discovery with recurrent neural networks
  publication-title: ACS Cent. Sci.
  doi: 10.1021/acscentsci.7b00512
– volume: 9
  year: 2017
  ident: 10.1016/j.ces.2025.122575_bib0047
  article-title: Molecular de-novo design through deep reinforcement learning
  publication-title: J. Cheminform.
  doi: 10.1186/s13321-017-0235-x
– volume: 13
  start-page: 1
  issue: 1
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0006
  article-title: Molecular generation by fast assembly of (deep)SMILES fragments
  publication-title: J. Cheminform.
  doi: 10.1186/s13321-021-00566-4
– start-page: 514
  year: 2023
  ident: 10.1016/j.ces.2025.122575_bib0039
  article-title: MoVAE: a variational autoencoder for molecular graph generation
– start-page: 7192
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0044
  article-title: Graphdf: a discrete flow model for molecular graph generation
– volume: 33
  start-page: 889
  issue: 12
  year: 2000
  ident: 10.1016/j.ces.2025.122575_bib0034
  article-title: Calculating structures and free energies of complex molecules: combining molecular mechanics and continuum models
  publication-title: Acc. Chem. Res.
  doi: 10.1021/ar000033j
– volume: 37
  start-page: 38559
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0070
  article-title: Generalized protein pocket generation with prior-informed flow matching
  publication-title: Adv. Neural Inf. Process. Syst.
– volume: 171
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0042
  article-title: Git-mol: a multi-modal large language model for molecular science with graph, image, and text
  publication-title: Comput. Biol. Medicine
  doi: 10.1016/j.compbiomed.2024.108073
– ident: 10.1016/j.ces.2025.122575_bib0020
  doi: 10.1021/acs.jcim.3c01964
– volume: 25
  start-page: 1157
  issue: 9
  year: 2004
  ident: 10.1016/j.ces.2025.122575_bib0061
  article-title: Development and testing of a general amber force field
  publication-title: J. Comput. Chem.
  doi: 10.1002/jcc.20035
– volume: 57
  start-page: 86
  issue: 4
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0049
  article-title: From understanding diseases to drug design: can artificial intelligence bridge the gap?
  publication-title: Artif. Intell. Rev.
  doi: 10.1007/s10462-024-10714-5
– volume: 17
  start-page: 24
  issue: 1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0056
  article-title: Entropy and the kullback-leibler divergence for bayesian networks: computational complexity and efficient implementation
  publication-title: Algorithms
  doi: 10.3390/a17010024
– volume: 23
  start-page: 1623
  issue: 16
  year: 2002
  ident: 10.1016/j.ces.2025.122575_bib0032
  article-title: Fast, efficient generation of high-quality atomic charges. AM1-BCC model: II. parameterization and validation
  publication-title: J. Comput. Chem.
  doi: 10.1002/jcc.10128
– volume: 155
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0026
  article-title: Online semi-supervised active learning ensemble classification for evolving imbalanced data streams
  publication-title: Appl. Soft Comput.
  doi: 10.1016/j.asoc.2024.111452
– volume: 20
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0003
  article-title: PDB2DAT: automating LAMMPS data file generation from PDB molecular systems using python, rdkit, and pysimm
  publication-title: Softw. Impacts
  doi: 10.1016/j.simpa.2024.100656
– year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0038
– volume: 4
  start-page: 899
  issue: 12
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0054
  article-title: Structure-based drug design with equivariant diffusion models
  publication-title: Nature Comput. Sci.
  doi: 10.1038/s43588-024-00737-x
– volume: 54
  start-page: 985
  issue: 1
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0043
  article-title: A novel belief tanimoto coefficient with its applications in multisource information fusion
  publication-title: Appl. Intell.
  doi: 10.1007/s10489-023-05217-9
– ident: 10.1016/j.ces.2025.122575_bib0048
– volume: 12
  start-page: 69812
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0005
  article-title: Advancements in generative AI: a comprehensive review of gans, gpt, autoencoders, diffusion model, and transformers
  publication-title: IEEE Access
  doi: 10.1109/ACCESS.2024.3397775
– ident: 10.1016/j.ces.2025.122575_bib0027
– ident: 10.1016/j.ces.2025.122575_bib0033
– volume: 40
  issue: 10
  year: 2021
  ident: 10.1016/j.ces.2025.122575_bib0035
  article-title: Generative adversarial networks for de novo molecular design
  publication-title: Mol. Inform.
  doi: 10.1002/minf.202100045
– volume: 12
  start-page: 57071
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0011
  article-title: An attention-based improved u-net neural network model for semantic segmentation of moving objects
  publication-title: IEEE Access
  doi: 10.1109/ACCESS.2024.3391249
– volume: 57
  start-page: 42
  issue: 2
  year: 2024
  ident: 10.1016/j.ces.2025.122575_bib0045
  article-title: Variational autoencoders for 3D data processing
  publication-title: Artif. Intell. Rev.
  doi: 10.1007/s10462-023-10687-x
– volume: 4
  start-page: 268
  year: 2016
  ident: 10.1016/j.ces.2025.122575_bib0024
  article-title: Automatic chemical design using a data-driven continuous representation of molecules
  publication-title: ACS Cent. Sci.
  doi: 10.1021/acscentsci.7b00572
SSID ssj0007710
Score 2.48727
Snippet •A novel RL-inspired framework combines VAE and latent-space diffusion for molecule generation.•Affinity and similarity constraints guide generation toward...
SourceID crossref
elsevier
SourceType Index Database
Publisher
StartPage 122575
SubjectTerms Affinity constraints
Genetic algorithm optimization
Latent space generation
Molecular generation
Title Reinforcement learning-inspired molecular generation with latent space diffusion and genetic algorithm optimization under affinity and similarity constraints
URI https://dx.doi.org/10.1016/j.ces.2025.122575
Volume 320
WOSCitedRecordID wos001578765800005&url=https%3A%2F%2Fcvtisr.summon.serialssolutions.com%2F%23%21%2Fsearch%3Fho%3Df%26include.ft.matches%3Dt%26l%3Dnull%26q%3D
hasFullText 1
inHoldings 1
isFullTextHit
isPrint
journalDatabaseRights – providerCode: PRVESC
  databaseName: ScienceDirect
  issn: 0009-2509
  databaseCode: AIEXJ
  dateStart: 19950101
  customDbUrl:
  isFulltext: true
  dateEnd: 99991231
  titleUrlDefault: https://www.sciencedirect.com
  omitProxy: false
  ssIdentifier: ssj0007710
  providerName: Elsevier
link http://cvtisr.summon.serialssolutions.com/2.0.0/link/0/eLvHCXMwtZ1Pb9MwFMCtsnGAw8RfMQaTD5yoUtWJE8fHCQ2NCU1oGtBb5NjOlKpNp66Z9mX4EnxCnu3YyTYmsQOXKHVtJ-r79fn52X4PoQ8q1prmikQiJTKiAi5C0TwSGdj3mpRE2C3_P76yk5N8NuPfRqPf_izM1YI1TX59zS_-q6ihDIRtjs4-QNyhUyiAexA6XEHscP0nwZ9qGwxVWr-fzwpxHtWNWVMH83LpE-Ka7Mm6A8C6Yxdgd0IT0DHwZzeZU9pLv1nZVLWhXRfnqzXUXY5XoGuW3SFOm0x3PRZVVTfGqrfOePgWHmI-SmODmlQULmpUCIzgYxXoPibiuBuRewdr7balBIZ_tnY9pZ4PuD5yXtwZFM3bvrTtfOGnuinFaujfiK1_w53wDDqbR2Co8aHOTuLpQOsSUEou_8qdAcH5JuYTULoT6D2d9HVvBt--NSiGrYp-F9y8gC4K00XhuniEtmOWchgMtg--HM6Ow_jPGJn6_H3mvf1aut1VeOs9_m4NDSycs2dop5ua4AOH1HM00s0L9HQQsPIl-nUDLnwHLhzgwj1c2MCFHVzYwoUDXBhQwR1cOMCFh3BhCxf2cNkWPVx4ANcr9P3z4dmno6hL8BFJMKs3USJVJqtMZVlZKcIUh9tSx4JUXIhSU5FkkvEqnjJFclnGZQymFJWZoirJqEiT12irWTX6DcJMlGZunIDpwWhVqZymFKbKJtADlzxRu-ij_6GLCxfHpbhXtLuIelEUHfbOwCwAq_ubvX3IM_bQk572d2hrs271e_RYXm3qy_V-x9QfkuO0OQ
linkProvider Elsevier
openUrl ctx_ver=Z39.88-2004&ctx_enc=info%3Aofi%2Fenc%3AUTF-8&rfr_id=info%3Asid%2Fsummon.serialssolutions.com&rft_val_fmt=info%3Aofi%2Ffmt%3Akev%3Amtx%3Ajournal&rft.genre=article&rft.atitle=Reinforcement+learning-inspired+molecular+generation+with+latent+space+diffusion+and+genetic+algorithm+optimization+under+affinity+and+similarity+constraints&rft.jtitle=Chemical+engineering+science&rft.au=Bai%2C+Can&rft.au=Wu%2C+Zijian&rft.au=Han%2C+Xianjun&rft.au=Huang%2C+Renbao&rft.date=2026-01-15&rft.issn=0009-2509&rft.volume=320&rft.spage=122575&rft_id=info:doi/10.1016%2Fj.ces.2025.122575&rft.externalDBID=n%2Fa&rft.externalDocID=10_1016_j_ces_2025_122575
thumbnail_l http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/lc.gif&issn=0009-2509&client=summon
thumbnail_m http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/mc.gif&issn=0009-2509&client=summon
thumbnail_s http://covers-cdn.summon.serialssolutions.com/index.aspx?isbn=/sc.gif&issn=0009-2509&client=summon