CoChat
Sign Up
Cited by 1,537Computational Drug Discovery MethodsMachine Learning in Materials ScienceSynthesis and Characterization of Heterocyclic Compounds

Generating Focused Molecule Libraries for Drug Discovery with Recurrent Neural Networks

Open access
Segler, Marwin H. S., Kogej, Thierry, Tyrchan, Christian, Waller, Mark P.·2018·ACS Central Science
DOI 10.1021/acscentsci.7b00512PMC PMC5785775
Open this paper in CoChat — free account

Abstract

High Resolution Image Download MS PowerPoint Slide In de novo drug design, computational strategies are used to generate novel molecules with good affinity to the desired biological target. In this work, we show that recurrent neural networks can be trained as generative models for molecular structures, similar to statistical language models in natural language processing. We demonstrate that the properties of the generated molecules correlate very well with the properties of the molecules used to train the model. In order to enrich libraries with molecules active toward a given biological target, we propose to fine-tune the model with small sets of molecules, which are known to be active against that target. Against Staphylococcus aureus, the model reproduced 14% of 6051 hold-out test molecules that medicinal chemists designed, whereas against Plasmodium falciparum (Malaria), it reproduced 28% of 1240 test molecules. When coupled with a scoring function, our model can perform the complete de novo drug design cycle to generate large sets of novel molecules for drug discovery.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1021/acscentsci.7b00512

Cites

64 of 64 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Claims other papers make about this one

Each is checked against this paper’s own text: does it say what the citing paper says it does?

  • Generative models have been used to compile focused and diversity-oriented chemical libraries, design inhibitors de novo, and perform hit-to-lead optimization, typically in a multiobjective mode.

    “Generative models have been used to compile both focused and diversity-oriented chemical libraries,149,150 to design inhibitors de novo,151 and to perform hit-to-lead optimization,152 typically in a multiobjective mode where e.g. selectivity, synthetic accessibility, and drug-like properties are improved simultaneously.153,154”

    From Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence (2023) · cites this paper as [149]

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 10

  1. Segler M. H.; Preuss M.; Waller M. P. Learning to Plan Chemical Syntheses. ArXiv 2017, 1708.04202.
  2. Duvenaud D. K.; Maclaurin D.; Iparraguirre J.; Bombarell R.; Hirzel T.; Aspuru-Guzik A.; Adams R. P. Convolutional networks on graphs for learning molecular fingerprints. Adv. Neural Inf. Proc. Syst. 2015, 2224– 2232.
  3. Jastrzebski S.; Lesniak D.; Czarnecki W. M. Learning to SMILE(S). In International Conference on Learning Representations; 2016.
  4. Unterthiner T.; Mayr A.; Klambauer G.; Steijaert M.; Wegner J. K.; Ceulemans H.; Hochreiter S. Deep learning as an opportunity in virtual screening. In Proceedings of the Deep Learning Workshop at NIPS; 2014; Vol. 27, pp 1– 9.
  5. Ramsundar B.; Kearnes S.; Riley P.; Webster D.; Konerding D.; Pande V. Massively multitask networks for drug discovery. ArXiv 2015, 1502.02072.
  6. Jozefowicz R.; Vinyals O.; Schuster M.; Shazeer N.; Wu Y. Exploring the limits of language modeling. ArXiv 2016, 1602.02410.
  7. Srivastava N.; Mansimov E.; Salakhutdinov R. Unsupervised learning of video representations using LSTMs. In Proceedings of the 32nd International Conference on Machine Learning; 2015; pp 843– 852.
  8. Bhoopchand A.; Rocktäschel T.; Barr E.; Riedel S. Learning Python Code Suggestion with a Sparse Pointer Network. ArXiv 2016, 1611.08307.
  9. Johnson M.; Schuster M.; Le Q. V.; Krikun M.; Wu Y.; Chen Z.; Thorat N.; Viégas F.; Wattenberg M.; Corrado G. Google’s Multilingual Neural Machine Translation System: Enabling Zero-Shot Translation. ArXiv 2016, 1611.04558.
  10. Graves A. Generating sequences with recurrent neural networks. ArXiv 2013, 1308.0850.

Cited by

Sorted by citations

Related

Authors
Segler, M. H. S., Kogej, T., Tyrchan, C., & Waller, M. P.
Affiliations
  • University of Münster (DE)
  • AstraZeneca (Sweden) (SE)
  • Shanghai University (CN)
Corresponding author
  • Segler, Marwin H. S.
  • Waller, Mark P.
ORCID
Venue
ACS Central Science
Publisher
American Chemical Society (ACS)
Published
2017-12-28
Volume
4
Issue
1
Pages
120-131
ISSN
  • 2374-7943 (ISSN-L)
  • 2374-7951
DOI
10.1021/acscentsci.7b00512
PMC
PMC5785775
OpenAlex
W2578240541
Cited by
1,537
References
64
Citations by year
2026: 114 · 2025: 163 · 2024: 208 · 2023: 218 · 2022: 183 · 2021: 257 · 2020: 178 · 2019: 130 · 2018: 75 · 2017: 10
FWCI
80.60
Type
journal article
Language
en
Open access
Licence
Locations
  • ACS Central Science (publishedVersion, other-oa)
  • PubMed Central (publishedVersion)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
Funders
  • Deutsche Forschungsgemeinschaft (SFB 858)
  • Shanghai Municipal Education Commission
Topics
  • Physical Sciences › Computer Science › Computational Theory and Mathematics › Computational Drug Discovery Methods
  • Physical Sciences › Materials Science › Materials Chemistry › Machine Learning in Materials Science
  • Physical Sciences › Chemistry › Organic Chemistry › Synthesis and Characterization of Heterocyclic Compounds
Keywords
de novo drug design, recurrent neural networks, fine-tuning, Staphylococcus aureus, Plasmodium falciparum, malaria, antibacterial agents, molecular property prediction, scoring functions

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review