Computational design of syntheses leading to compound libraries or isotopically labelled targets

Molga, Karol; Dittwald, Piotr; Grzybowski, Bartosz A.

doi:10.1039/c9sc02678a

Cited by 19 publications

(20 citation statements)

References 79 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Using intelligent methods to choose a means by which to make them oneself is entirely another [152][153][154][157][158][159][160][161][162]. Probably the present apotheosis of this strategy is Chematica [157,158,163] (recently commercialised as Synthia TM ), that started life as a manually encoded set of reaction rules and now uses (so far as one can tell from published material) a variety of means of intelligent search. The exciting prospect is for the discovery of entirely novel reactions based on those presently known; this is precisely the area in which generative methods can excel.…”

Section: Chemical Synthesesmentioning

confidence: 99%

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

Kell

Samanta

Swainston

2020

Biochemical Journal

View full text Add to dashboard Cite

The number of ‘small’ molecules that may be of interest to chemical biologists — chemical space — is enormous, but the fraction that have ever been made is tiny. Most strategies are discriminative, i.e. have involved ‘forward’ problems (have molecule, establish properties). However, we normally wish to solve the much harder generative or inverse problem (describe desired properties, find molecule). ‘Deep’ (machine) learning based on large-scale neural networks underpins technologies such as computer vision, natural language processing, driverless cars, and world-leading performance in games such as Go; it can also be applied to the solution of inverse problems in chemical biology. In particular, recent developments in deep learning admit the in silico generation of candidate molecular structures and the prediction of their properties, thereby allowing one to navigate (bio)chemical space intelligently. These methods are revolutionary but require an understanding of both (bio)chemistry and computer science to be exploited to best advantage. We give a high-level (non-mathematical) background to the deep learning revolution, and set out the crucial issue for chemical biology and informatics as a two-way mapping from the discrete nature of individual molecules to the continuous but high-dimensional latent representation that may best reflect chemical space. A variety of architectures can do this; we focus on a particular type known as variational autoencoders. We then provide some examples of recent successes of these kinds of approach, and a look towards the future.

show abstract

Section: Chemical Synthesesmentioning

confidence: 99%

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

Kell

Samanta

Swainston

2020

Biochemical Journal

View full text Add to dashboard Cite

show abstract

“…We are not taking here about syntheses planned one-by-one but about "global plans" that make use of intermediates and starting materials common to multiple individual pathways (the use of such common intermediates/substrates may lower the overall cost of the process). As described in ref 30 , Chematica is quite adept in constructing such plans within minutes to hours; in the example in Figure 7 the software's task was to design a global synthetic plan leading to the synthetically most accessible M+6, 13 C isotopically labelled derivatives of ten anticoagulant rodenticides. Note how intricate this plan is -it looks like a small network, not just a synthetic path.…”

Section: Synthesis With Multiple Constraintsmentioning

confidence: 99%

“…Note how intricate this plan is -it looks like a small network, not just a synthetic path. The problem of library-wide design has many more ramifications (e.g., in ranking library members for the ease of synthesis, or in the selection of the most synthetically accessible isotopomers, for details see 30 ). These tasks definitely require computer's assistance when the numbers of library members or the ways in which a compound can be multiply labelled become large (for details, see 30 ).…”

Section: Synthesis With Multiple Constraintsmentioning

confidence: 99%

Engines of discovery: Computers in advanced synthesis planning and identification of drug candidates

Grzybowski¹

2020

ACM

Self Cite

View full text Add to dashboard Cite

show abstract

“…Recent advances in computing technology and artificial intelligence (AI) have enabled significant progress in the decades‐old problem of teaching computers the planning of chemical syntheses, and have resulted in the first set of experimental validations of complete, computer‐designed synthetic plans leading to non‐trivial targets . An integral portion of software platforms such as our Chematica,– 3 N‐MCTS by Waller, ASKCOS by the MIT, or RXN by IBM are the so‐called scoring functions (SFs) that evaluate individual “synthetic moves” and guide the navigation over the vast networks of synthetic possibilities, ultimately seeking commercially available starting materials. These scoring functions have been either expert‐defined or machine‐trained.…”

Section: Figurementioning

confidence: 99%

Synergy Between Expert and Machine‐Learning Approaches Allows for Improved Retrosynthetic Planning

et al. 2019

Self Cite

View full text Add to dashboard Cite

When computers plan multistep syntheses, they can rely either on expert knowledge or information machine‐extracted from large reaction repositories. Both approaches suffer from imperfect functions evaluating reaction choices: expert functions are heuristics based on chemical intuition, whereas machine learning (ML) relies on neural networks (NNs) that can make meaningful predictions only about popular reaction types. This paper shows that expert and ML approaches can be synergistic—specifically, when NNs are trained on literature data matched onto high‐quality, expert‐coded reaction rules, they achieve higher synthetic accuracy than either of the methods alone and, importantly, can also handle rare/specialized reaction types.

show abstract

Computational design of syntheses leading to compound libraries or isotopically labelled targets

Abstract: Network-search routines over large graphs of retrosynthetic scenarios are adapted to multi-target design operating on one common search graph enabling design of syntheses of compound libraries or isotopically labelled targets.

Cited by 19 publications

References 79 publications

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

Deep learning and generative methods in cheminformatics and chemical biology: navigating small molecule space intelligently

Engines of discovery: Computers in advanced synthesis planning and identification of drug candidates

Synergy Between Expert and Machine‐Learning Approaches Allows for Improved Retrosynthetic Planning

Contact Info

Product

Resources

About