Molecule Ideation Using Matched Molecular Pairs

Pal, Siladitya; Pogány, Péter; Lumley, James A.

doi:10.1007/978-1-0716-1787-8_23

Cited by 4 publications

(6 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Since the transformations shown in Figure , which produce highly active molecules against a given protein target, could be identified with a traditional matched molecular pairs (MMP) approach, ,, we sought to evaluate the ability of the proposed approach to go beyond MMP . Therefore, we first generated all MMP transformation rules (in the form of SMIRKS) for our training subset of ChEMBL molecules; expectedly, ,, the most ubiquitous of them were additions or replacements of single atoms (H, F, Cl, etc.) or simple groups (methyl, methoxy, ethyl, etc.)…”

Section: Resultsmentioning

confidence: 99%

“…So far the most popular computational algorithm for lead optimization has arguably been MMP analysis . ,,− This method is based on extraction of transformation rules from a training set of molecules, and application of these rules to known actives. Transformations captured in this way are generic, the method is based on the additivity principle, and yields multiple suggested molecules.…”

Section: Discussionmentioning

confidence: 99%

See 1 more Smart Citation

Can We Quickly Learn to “Translate” Bioactive Molecules with Transformer Models?

Tysinger

Brajesh

Sinitskiy

2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Meaningful exploration of the chemical space of druglike molecules in drug design is a highly challenging task due to a combinatorial explosion of possible modifications of molecules. In this work, we address this problem with transformer models, a type of machine learning (ML) model originally developed for machine translation. By training transformer models on pairs of similar bioactive molecules from the public ChEMBL data set, we enable them to learn medicinal-chemistry-meaningful, context-dependent transformations of molecules, including those absent from the training set. By retrospective analysis on the performance of transformer models on ChEMBL subsets of ligands binding to COX2, DRD2, or HERG protein targets, we demonstrate that the models can generate structures identical or highly similar to most active ligands, despite the models having not seen any ligands active against the corresponding protein target during training. Our work demonstrates that human experts working on hit expansion in drug design can easily and quickly employ transformer models, originally developed to translate texts from one natural language to another, to "translate" from known molecules active against a given protein target to novel molecules active against the same target.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Discussionmentioning

confidence: 99%

Can We Quickly Learn to “Translate” Bioactive Molecules with Transformer Models?

Tysinger

Brajesh

Sinitskiy

2023

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…First of all, we compare our approach to MMP analysis, 3,4,16,19 a popular algorithm for lead optimization. This method is based on extraction of transformation rules from (in the terminology of ML) a training set of molecules, and application of these rules to known actives.…”

Section: Discussionmentioning

confidence: 99%

“…In this work, we show that transformer models are not only able to generate new molecular structures absent from the training dataset, but, in doing so, also go beyond the standard matched molecular pairs (MMP)-based approach. 3,4,16 To demonstrate this, we first generated all MMP transformation rules (in the form of SMIRKS) for our training subset of ChEMBL 15 molecules; expectedly, 4,16 the most ubiquitous of them were additions or replacements of single atoms (H, F, Cl, etc.) or simple groups (methyl, methoxy, ethyl, etc.)…”

Section: Transformer Models Trained On Pairs Of Bioactive Molecules C...mentioning

confidence: 99%

Can We Quickly Learn to “Translate” Bioactive Molecules with Transformer Models?

Tysinger

Brajesh

Sinitskiy

2022

Preprint

View full text Add to dashboard Cite

Meaningful exploration of the chemical space of druglike molecules in drug design is a highly challenging task due to a combinatorial explosion of possible modifications of molecules. In this work, we address this problem with transformer models, a type of machine learning (ML) model, with recent demonstrated success in applications to machine translation and other tasks. By training transformer models on pairs of similar bioactive molecules from the public ChEMBL dataset, we enable them to learn medicinal-chemistry-meaningful, context-dependent transformations of molecules, including those absent from the training set. Most generated molecules are highly plausible and follow similar distributions of simple properties (molecular weight, polarity, hydrogen bond donor and acceptor numbers) as the training dataset. By retrospective analysis of the performance of transformer models on ChEMBL subsets of ligands binding to COX2, DRD2, or HERG protein targets, we demonstrate that the models can generate structures identical or highly similar to highly active ligands, despite the models having not seen any ligands active against the corresponding protein target during training. Thus, our work demonstrates that transformer models, originally developed to translate texts from one natural language to another, can be easily and quickly extended to “translations” from known molecules active against a given protein target to novel molecules active against the same target, and thereby contribute to hit expansion in drug design.

show abstract

“…Traditional strategies, such as scaffold hopping and bioisosteric replacement, mainly depend on the empirical chemical rules from medicinal chemists. , However, given the rising amount of data and complexity of chemical and biological systems, it is getting more difficult for medicinal chemists to manually extract related chemical rules . Accordingly, researchers developed many computational methods to automatically learn hidden medicinal chemistry knowledge from large data sets for the prediction and optimization of pharmacokinetic properties, such as machine learning-based quantitative structure–activity relationship (QSAR) models − and matched molecular pairs analysis (MMPA). − However, these strategies have certain limitations. For example, QSAR models cannot provide structural modification schemes, and MMPA can only be used to extract practical transformations from molecules sharing the same context .…”

Section: Introductionmentioning

confidence: 99%

IDL-PPBopt: A Strategy for Prediction and Optimization of Human Plasma Protein Binding of Compounds via an Interpretable Deep Learning Method

Lou

Yang

Wang

et al. 2022

J. Chem. Inf. Model.

View full text Add to dashboard Cite

The prediction and optimization of pharmacokinetic properties are essential in lead optimization. Traditional strategies mainly depend on the empirical chemical rules from medicinal chemists. However, with the rising amount of data, it is getting more difficult to manually extract useful medicinal chemistry knowledge. To this end, we introduced IDL-PPBopt, a computational strategy for predicting and optimizing the plasma protein binding (PPB) property based on an interpretable deep learning method. At first, a curated PPB data set was used to construct an interpretable deep learning model, which showed excellent predictive performance with a root mean squared error of 0.112 for the entire test set. Then, we designed a detection protocol based on the model and Wilcoxon test to identify the PPB-related substructures (named privileged substructures, PSubs) for each molecule. In total, 22 general privileged substructures (GPSubs) were identified, which shared some common features such as nitrogen-containing groups, diamines with two carbon units, and azetidine. Furthermore, a series of second-level chemical rules for each GPSub were derived through a statistical test and then summarized into substructure pairs. We demonstrated that these substructure pairs were equally applicable outside the training set and accordingly customized the structural modification schemes for each GPSub, which provided alternatives for the optimization of the PPB property. Therefore, IDL-PPBopt provides a promising scheme for the prediction and optimization of the PPB property and would be helpful for lead optimization of other pharmacokinetic properties.

show abstract

Molecule Ideation Using Matched Molecular Pairs

Cited by 4 publications

References 18 publications

Can We Quickly Learn to “Translate” Bioactive Molecules with Transformer Models?

Can We Quickly Learn to “Translate” Bioactive Molecules with Transformer Models?

Can We Quickly Learn to “Translate” Bioactive Molecules with Transformer Models?

IDL-PPBopt: A Strategy for Prediction and Optimization of Human Plasma Protein Binding of Compounds via an Interpretable Deep Learning Method

Contact Info

Product

Resources

About