Coping with Combinatorial Space in Molecular Design

Lauck, Florian; Rarey, Matthias

doi:10.1002/9783527677016.ch14

Cited by 4 publications

(5 citation statements)

References 72 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…7 Efficient approaches to query larger combinatorial spaces without full enumeration via similarity to given structures have been reported. [8][9][10] However, it is not straightforward to perform more complex, multi-objective queries without enumerating substantial parts of the space.…”

Section: Introductionmentioning

confidence: 99%

“…However, this is only a tiny fraction of drug-like chemical space, which has an estimated size anywhere between 10 24 and 10 60 possible structures. , This number might be even larger considering new modalities such as PROTACs . Efficient approaches to query larger combinatorial spaces without full enumeration via similarity to given structures have been reported. − However, it is not straightforward to perform more complex, multiobjective queries without enumerating substantial parts of the space.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

GuacaMol: Benchmarking Models for de Novo Molecular Design

Brown

Fiscato

Segler

et al. 2019

J. Chem. Inf. Model.

634

883

View full text Add to dashboard Cite

De novo design seeks to generate molecules with required property profiles by virtual design-make-test cycles. With the emergence of deep learning and neural generative models in many application areas, models for molecular design based on neural networks appeared recently and show promising results. However, the new models have not been profiled on consistent tasks, and comparative studies to well-established algorithms have only seldom been performed. To standardize the assessment of both classical and neural models for de novo molecular design, we propose an evaluation framework, GuacaMol, based on a suite of standardized benchmarks. The benchmark tasks encompass measuring the fidelity of the models to reproduce the property distribution of the training sets, the ability to generate novel molecules, the exploration and exploitation of chemical space, and a variety of single and multi-objective optimization tasks. The benchmarking open-source Python code, and a leaderboard can be found on https://benevolent.ai/guacamol.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

GuacaMol: Benchmarking Models for de Novo Molecular Design

Brown

Fiscato

Segler

et al. 2019

J. Chem. Inf. Model.

634

883

View full text Add to dashboard Cite

show abstract

“…Fragment spaces are an efficient and elegant way to describe such a virtual, combinatorial library [8,9]. They consist of a number of fragments with pre-defined Christian Meyenburg and Uschi Dolfus have contributed equally to this work.…”

Section: Introductionmentioning

confidence: 99%

Galileo: Three-dimensional searching in large combinatorial fragment spaces on the example of pharmacophores

Meyenburg

Dolfus

Briem

et al. 2022

J Comput Aided Mol Des

Self Cite

View full text Add to dashboard Cite

Fragment spaces are an efficient way to model large chemical spaces using a handful of small fragments and a few connection rules. The development of Enamine’s REAL Space has shown that large spaces of readily available compounds may be created this way. These are several orders of magnitude larger than previous libraries. So far, searching and navigating these spaces is mostly limited to topological approaches. A way to overcome this limitation is optimization via metaheuristics which can be combined with arbitrary scoring functions. Here we present Galileo, a novel Genetic Algorithm to sample fragment spaces. We showcase Galileo in combination with a novel pharmacophore mapping approach, called Phariety, enabling 3D searches in fragment spaces. We estimate the effectiveness of the approach with a small fragment space. Furthermore, we apply Galileo to two pharmacophore searches in the REAL Space, detecting hundreds of compounds fulfilling a HSP90 and a FXIa pharmacophore.

show abstract

“…Fragment spaces can surpass noncombinatorial databases in the number of compounds contained by several orders of magnitude while requiring only a manageable amount of storage space. 20,21 However, common molecular descriptors can be inappropriate for the application to fragment spaces, and new methods tailored to the special conditions present in combinatorial libraries of small fragments are highly desirable. 22,23 We will show that the limitation to a structural feature family can be problematic for a fingerprint method when applied to fragments.…”

Section: ■ Introductionmentioning

confidence: 99%

Connected Subgraph Fingerprints: Representing Molecules Using Exhaustive Subgraph Enumeration

Bellmann

Penner

Rarey

2019

J. Chem. Inf. Model.

Self Cite

View full text Add to dashboard Cite

Molecular fingerprints are an efficient and widely used method for similarity-driven virtual screening. Most fingerprint methods can be distinguished by the class of structural features considered. The Connected Subgraph Fingerprint (CSFP) overcomes this limitation and regards all structural features of a compound. This results in a more complete feature space and high adaptive potential to certain application scenarios. The novel descriptor surpasses widely used fingerprint methods in some cases and opens the way for topological search in combinatorial fragment spaces.

show abstract

Coping with Combinatorial Space in Molecular Design

Cited by 4 publications

References 72 publications

GuacaMol: Benchmarking Models for de Novo Molecular Design

GuacaMol: Benchmarking Models for de Novo Molecular Design

Galileo: Three-dimensional searching in large combinatorial fragment spaces on the example of pharmacophores

Connected Subgraph Fingerprints: Representing Molecules Using Exhaustive Subgraph Enumeration

Contact Info

Product

Resources

About