DeepAC – conditional transformer-based chemical language model for the prediction of activity cliffs formed by bioactive compounds

Chen, Hengwei; Vogt, Martin; Bajorath, Jürgen

doi:10.1039/d2dd00077f

Cited by 13 publications

(17 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Previously, we have reported a deep learning approach for the prediction of ACs that further extended other ML classification methods by its ability to not only predict ACs, but also generate new AC compounds 17 . Since ACs encode large potency differences, we have reasoned that this methodology might be adapted and further extended for the design of highly potent compounds.…”

Section: Introductionmentioning

confidence: 99%

Designing highly potent compounds using a chemical language model

Chen

Bajorath

2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

Compound potency prediction is a major task in medicinal chemistry and drug design. Inspired by the concept of activity cliffs (which encode large differences in potency between similar active compounds), we have devised a new methodology for predicting potent compounds from weakly potent input molecules. Therefore, a chemical language model was implemented consisting of a conditional transformer architecture for compound design guided by observed potency differences. The model was evaluated using a newly generated compound test system enabling a rigorous assessment of its performance. It was shown to predict known potent compounds from different activity classes not encountered during training. Moreover, the model was capable of creating highly potent compounds that were structurally distinct from input molecules. It also produced many novel candidate compounds not included in test sets. Taken together, the findings confirmed the ability of the new methodology to generate structurally diverse highly potent compounds.

show abstract

Section: Introductionmentioning

confidence: 99%

Designing highly potent compounds using a chemical language model

Chen

Bajorath

2023

Sci Rep

Self Cite

View full text Add to dashboard Cite

show abstract

“…These DL approaches to AC prediction reached similarly high prediction accuracy as earlier ML studies (for example, with area under the receiver-operating characteristic curve (AUC) values greater 0.9). Furthermore, a transformer-based chemical language model has recently been introduced to bridge between AC prediction and the design of new AC compounds [ 13 ], hence adding a new dimension to predictive modeling. This model also achieved AC prediction accuracy comparable to (or better than) other state-of-the-art ML models [ 13 ].…”

Section: Introductionmentioning

confidence: 99%

“…Furthermore, a transformer-based chemical language model has recently been introduced to bridge between AC prediction and the design of new AC compounds [ 13 ], hence adding a new dimension to predictive modeling. This model also achieved AC prediction accuracy comparable to (or better than) other state-of-the-art ML models [ 13 ]. In addition to classification models for AC prediction, regression models have also been applied to predict the potency of individual AC compounds [ 14 , 15 ].…”

Section: Introductionmentioning

confidence: 99%

Large-scale prediction of activity cliffs using machine and deep learning methods of increasing complexity

2023

Self Cite

View full text Add to dashboard Cite

Activity cliffs (AC) are formed by pairs of structural analogues that are active against the same target but have a large difference in potency. While much of our knowledge about ACs has originated from the analysis and comparison of compounds and activity data, several studies have reported AC predictions over the past decade. Different from typical compound classification tasks, AC predictions must be carried out at the level of compound pairs representing ACs or nonACs. Most AC predictions reported so far have focused on individual methods or comparisons of two or three approaches and only investigated a few compound activity classes (from 2 to 10). Although promising prediction accuracy has been reported in most cases, different system set-ups, AC definitions, methods, and calculation conditions were used, precluding direct comparisons of these studies. Therefore, we have carried out a large-scale AC prediction campaign across 100 activity classes comparing machine learning methods of greatly varying complexity, ranging from pair-based nearest neighbor classifiers and decision tree or kernel methods to deep neural networks. The results of our systematic predictions revealed the level of accuracy that can be expected for AC predictions across many different compound classes. In addition, prediction accuracy did not scale with methodological complexity but was significantly influenced by memorization of compounds shared by different ACs or nonACs. In many instances, limited training data were sufficient for building accurate models using different methods and there was no detectable advantage of deep learning over simpler approaches for AC prediction. On a global scale, support vector machine models performed best, by only small margins compared to others including simple nearest neighbor classifiers. Graphical Abstract

show abstract

“…The AC-prediction literature is still very thin compared to the QSAR-prediction literature. An attempt to conduct an exhaustive literature review on AC-prediction techniques revealed a total number of 15 methods [ 3 , 5 , 7 , 11 , 19 , 21 , 24 , 26 , 30 , 34 , 41 – 43 , 46 , 57 ], all of which have been published since 2012. Current AC-prediction methods are often based on creative ways to extract features from pairs of molecular compounds in a manner suitable for standard machine learning pipelines.…”

Section: Introductionmentioning

confidence: 99%

Exploring QSAR models for activity-cliff prediction

et al. 2023

View full text Add to dashboard Cite

Introduction and methodology Pairs of similar compounds that only differ by a small structural modification but exhibit a large difference in their binding affinity for a given target are known as activity cliffs (ACs). It has been hypothesised that QSAR models struggle to predict ACs and that ACs thus form a major source of prediction error. However, the AC-prediction power of modern QSAR methods and its quantitative relationship to general QSAR-prediction performance is still underexplored. We systematically construct nine distinct QSAR models by combining three molecular representation methods (extended-connectivity fingerprints, physicochemical-descriptor vectors and graph isomorphism networks) with three regression techniques (random forests, k-nearest neighbours and multilayer perceptrons); we then use each resulting model to classify pairs of similar compounds as ACs or non-ACs and to predict the activities of individual molecules in three case studies: dopamine receptor D2, factor Xa, and SARS-CoV-2 main protease. Results and conclusions Our results provide strong support for the hypothesis that indeed QSAR models frequently fail to predict ACs. We observe low AC-sensitivity amongst the evaluated models when the activities of both compounds are unknown, but a substantial increase in AC-sensitivity when the actual activity of one of the compounds is given. Graph isomorphism features are found to be competitive with or superior to classical molecular representations for AC-classification and can thus be employed as baseline AC-prediction models or simple compound-optimisation tools. For general QSAR-prediction, however, extended-connectivity fingerprints still consistently deliver the best performance amongs the tested input representations. A potential future pathway to improve QSAR-modelling performance might be the development of techniques to increase AC-sensitivity. Graphical Abstract

show abstract

DeepAC – conditional transformer-based chemical language model for the prediction of activity cliffs formed by bioactive compounds

Abstract: Activity cliffs (ACs) are formed by pairs of structurally similar or analogous active small molecules with large differences in potency. In medicinal chemistry, ACs are of high interest because they...

Cited by 13 publications

References 34 publications

Designing highly potent compounds using a chemical language model

Designing highly potent compounds using a chemical language model

Large-scale prediction of activity cliffs using machine and deep learning methods of increasing complexity

Exploring QSAR models for activity-cliff prediction

Contact Info

Product

Resources

About