MétaCan
Menu
Back to cohort
Record W4416395717 · doi:10.1021/acs.jcim.5c02234

Generative AI for the Design of Molecules: Advances and Challenges

2025· article· en· W4416395717 on OpenAlexafffund
Yan Sun, Lianghong Chen, Zihao Jing, Yan Yi Li, Dongkyu Kim, Reza Noroozi, Grace Y. Yi, Conrard Giresse Tetsassi Feugmo, Anna Klinkova, Kyla N. Sask, Agustinus Kristiadi, Boyu Wang, Elizabeth R. Gillies, Kun Ping Lu, Haotian Shi, Pingzhao Hu

Bibliographic record

VenueJournal of Chemical Information and Modeling · 2025
Typearticle
Languageen
FieldComputer Science
TopicComputational Drug Discovery Methods
Canadian institutionsMcMaster UniversityUniversity of WaterlooWestern University
FundersCanadian Institutes of Health ResearchNatural Sciences and Engineering Research Council of CanadaCanada Research ChairsCanada Foundation for Innovation
KeywordsGenerative grammarGenerative modelChemical spaceGenerative DesignDrug discoveryParadigm shiftDeep learning

Abstract

fetched live from OpenAlex

The design of novel molecules underpins advances in both drug discovery and biomaterials engineering. Traditional approaches, from natural product isolation to high-throughput screening, have delivered important therapeutics but remain costly, inefficient, and limited in exploring the chemical and biomolecular space. While predictive machine learning models improved aspects of discovery, they cannot fully address the complexity of modern precision medicine. Generative artificial intelligence (AI) offers a paradigm shift by enabling de novo molecular creation guided by data-driven optimization. Architectures such as variational autoencoders, generative adversarial networks, normalizing flows, and diffusion models now demonstrate unprecedented capabilities in designing small molecules and macromolecules that satisfy complex physicochemical and biological requirements. This review surveys the rapidly evolving field of generative AI for molecular design. We first introduce the development of generative architectures and optimization strategies, focusing on how sampling, training, and postgeneration techniques improve control over molecular design. We then examine applications across molecular representations, unconstrained and property-constrained design, conformation modeling, and the generation of large biomolecules such as proteins, antibodies, and peptides. Benchmarking datasets, evaluation metrics, and real-world case studies, such as the AI-driven discovery of novel antibiotics demonstrated in vivo efficacy against multidrug-resistant infections, illustrate the growing maturity and translational potential of generative molecular design approaches. Despite rapid advances, generative molecular design still faces critical challenges that point to key future directions. These include integrating physicochemical priors through differentiable physical models, overcoming data scarcity via synthetic augmentation and transfer learning, enabling multimodal fusion of structural, omics, and phenotypic data, deploying autonomous AI agents for adaptive decision-making, and optimizing multiple objectives with uncertainty-aware strategies. Addressing these challenges could lead to more robust, generalizable, and experimentally aligned molecular design systems.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.

How this classification was reachedexpand

Full frame distilled prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.

metaresearch head score (Codex)0.000
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: Simulation or modeling
GenreCandidate signal: Methods · Consensus signal: Methods
Teacher disagreement score0.419
Threshold uncertainty score0.104

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.001
Open science0.0000.000
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

Machine scores (provisional)

The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.

Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.

Opus teacher head0.066
GPT teacher head0.341
Teacher spread0.275 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it

Classification

machine, unvalidated

Machine predicted; a candidate call from one teacher head, not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designSimulation or modeling
Domainnot available
GenreMethods

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations7
Published2025
Admission routes2
Has abstractyes

Explore more

Same venueJournal of Chemical Information and ModelingSame topicComputational Drug Discovery MethodsFrench-language works237,207