The (Re)-Evolution of Quantitative Structure–Activity Relationship (QSAR) Studies Propelled by the Surge of Machine Learning Methods

Soares, Thereza A.; Nunes-Alves, Ariane; Mazzolari, Angelica; Ruggiu, Fiorella; Wei, Guo-Wei; Merz, Kenneth M.

doi:10.1021/acs.jcim.2c01422

Cited by 37 publications

(28 citation statements)

References 47 publications

(74 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…This equation was formed by Hansch and Fujita, and it uses ED50 as the activity parameter and the electrical parameter., steric parameters, and hydrophobic parameters as variables for linear regression analysis. 6 Guided by the Hansch equation, 4-quinolone antibacterial drugs such as norfloxacin have been successfully designed, and it proves the validity of the Hansch equation. 7 However, the Hansch equation has many parameters that make the modeling process difficult.…”

Section: Introductionmentioning

confidence: 78%

“…The most common approach for the prediction of biological activity based on ligands is the quantitative structure–activity relationship (QSAR) proposed by Hansch et al, , which is based on the principle that ligand activity is correlated with molecular structure and the activity value can be predicted by establishing a mathematical model based on the molecular structure of ligands. , The Hansch equation is the first one implemented in QSAR. This equation was formed by Hansch and Fujita, and it uses ED50 as the activity parameter and the electrical parameter., steric parameters, and hydrophobic parameters as variables for linear regression analysis . Guided by the Hansch equation, 4-quinolone antibacterial drugs such as norfloxacin have been successfully designed, and it proves the validity of the Hansch equation .…”

Section: Introductionmentioning

confidence: 79%

“…Data Sets. Six data sets are studied in this paper: (1) angiotensin-converting enzyme (ACE) inhibitors, which are zinc-containing metallopeptidases; 21 (2) acetylcholinesterase (AchE) inhibitors, which can bring about memory improvement in SDAT; 16,22 (3) benzodiazepine receptors (BZRs), which are commonly used in the therapeutic treatment of anxiety, insomnia, seizure disorders, and certain kinds of spasticity; 23,24 (4) cyclooxygenase-2 (COX2), which is a highly selective and potent inhibitor; 25,26 (5) dihydrofolate reductase (DHFR) inhibitors, which can produce unacceptable toxicity to host proliferative tissues; 27−29 (6) estrogen receptor alpha (ERα) inhibitors, which are considered possible candidates for the treatment of breast cancer. 30 Representative structures of the compounds are shown in Figure 1.…”

Section: Data Sets and Modelsmentioning

confidence: 99%

See 2 more Smart Citations

Biological Activity Predictions of Ligands Based on Hybrid Molecular Fingerprinting and Ensemble Learning

Zeng

Zhang

et al. 2023

ACS Omega

View full text Add to dashboard Cite

The biological activity predictions of ligands are an important research direction, which can improve the efficiency and success probability of drug screening. However, the traditional prediction method has the disadvantages of complex modeling and low screening efficiency. Machine learning is considered an important research direction to solve these traditional method problems in the near future. This paper proposes a machine learning model with high predictive accuracy and stable prediction ability, namely, the back propagation neural network cross-support vector regression model (BPCSVR). By comparing multiple molecular descriptors, MACCS fingerprint and ECFP6 fingerprint were selected as inputs, and the stable prediction ability of the model was improved by integrating multiple models and correcting similar samples. We used leave-one-out cross-validation on 3038 samples from six data sets. The coefficient of determination, root mean square error, and absolute error were used as the evaluation parameters. After comparing the multiclass models, the results show that the BPCSVR model has stable prediction ability in different data sets, and the prediction accuracy is higher than other comparison models.

show abstract

Section: Introductionmentioning

confidence: 78%

Section: Introductionmentioning

confidence: 79%

Section: Data Sets and Modelsmentioning

confidence: 99%

See 1 more Smart Citation

Biological Activity Predictions of Ligands Based on Hybrid Molecular Fingerprinting and Ensemble Learning

Zeng

Zhang

et al. 2023

ACS Omega

View full text Add to dashboard Cite

show abstract

“…Emerging from pattern recognition studies and the concept of computational learning, ML algorithms can adapt and update during the process of training without explicit programming to do so–in turn improving predictive accuracy in an automated manner [ 7 ]. As such, they are now identified as one of the most vital and rapidly evolving areas in chemoinformatics [ 8 , 9 ]. Broadly, two overarching classes of ML may be distinguished: supervised or unsupervised.…”

Section: Introductionmentioning

confidence: 99%

Guidance for good practice in the application of machine learning in development of toxicological quantitative structure-activity relationships (QSARs)

et al. 2023

View full text Add to dashboard Cite

Recent years have seen a substantial growth in the adoption of machine learning approaches for the purposes of quantitative structure-activity relationship (QSAR) development. Such a trend has coincided with desire to see a shifting in the focus of methodology employed within chemical safety assessment: away from traditional reliance upon animal-intensive in vivo protocols, and towards increased application of in silico (or computational) predictive toxicology. With QSAR central amongst techniques applied in this area, the emergence of algorithms trained through machine learning with the objective of toxicity estimation has, quite naturally, arisen. On account of the pattern-recognition capabilities of the underlying methods, the statistical power of the ensuing models is potentially considerable–appropriate for the handling even of vast, heterogeneous datasets. However, such potency comes at a price: this manifesting as the general practical deficits observed with respect to the reproducibility, interpretability and generalisability of the resulting tools. Unsurprisingly, these elements have served to hinder broader uptake (most notably within a regulatory setting). Areas of uncertainty liable to accompany (and hence detract from applicability of) toxicological QSAR have previously been highlighted, accompanied by the forwarding of suggestions for “best practice” aimed at mitigation of their influence. However, the scope of such exercises has remained limited to “classical” QSAR–that conducted through use of linear regression and related techniques, with the adoption of comparatively few features or descriptors. Accordingly, the intention of this study has been to extend the remit of best practice guidance, so as to address concerns specific to employment of machine learning within the field. In doing so, the impact of strategies aimed at enhancing the transparency (feature importance, feature reduction), generalisability (cross-validation) and predictive power (hyperparameter optimisation) of algorithms, trained upon real toxicity data through six common learning approaches, is evaluated.

show abstract

“…However, conventional ML approaches for QSAR modeling have mainly focused so far on feature engineering for molecular descriptors based on fingerprint-, InChi-, SMILES-or 2Dgraph-based molecular representations. 6,[8][9][10][11] Besides the remarkable results of QSAR models in the past side by side with the improvements in virtual screening, there is evidence that the 3D structure of the molecules can significantly influence physical, chemical, and biological activity. 4,12,13 For instance, cis-Platin is used as a chemotherapy drug, whereas its stereoisomer, trans-Platin, does not show cytotoxic activity.…”

Section: Introductionmentioning

confidence: 99%

Equivariant Graph Neural Networks for Toxicity Prediction

Cremer

Sandonas

Tkatchenko

et al. 2023

Preprint

View full text Add to dashboard Cite

Predictive modeling for toxicity is a crucial step in the drug discovery pipeline. It can help filter out molecules with a high probability of failing in the early stages of de novo drug design. Thus, several machine learning (ML) models have been developed to predict the toxicity of molecules by combining classical ML techniques or deep neural networks with well-known molecular representations such as fingerprints, InChi keys, SMILES strings, or 2D-graphs. Since molecules live in the physical 3D space, it is expected that their more natural and accurate representation would be defined in this geometric space as in ab initio methods for computing physicochemical properties. In such space, these molecules are also subject to geometric symmetries that only adequate 3D molecular representations can capture. Hence, the implications of using 3D structures in toxicity predictive modeling is an essential yet open question that needs further investigation. Recent studies showed great success in using equivariant graph neural networks (EGNNs) for representation learning based on 3D structures to predict potential energy surfaces and physicochemical properties of organic molecules and materials. Inspired by this, we investigate the performance of EG-NNs to construct reliable ML models for toxicity prediction. We use the equivariant transformer (ET) model in TorchMD-NET, an SE(3)-equivariant attention-weighted message-passing neural network. Eleven different toxicity datasets taken from Molecu-leNet, Therapeutic Data Commons (TDC), and ToxBenchmark have been considered to evaluate the capability of ET for toxicity prediction. Our results show that ET adequately learns 3D representations of molecules that can be successfully correlated with toxicity activity, achieving good accuracies on most datasets comparable to state-of-the-art 2D graph-based models and significantly better than a SMILES-based trans-former model. However, exhaustive (single- and multi-) conformational analysis has revealed a lack of correlation between the conformational samples and the predictive performance, which we find is only slightly affected by the conformer selection. Moreover, combining toxicity prediction in a multi-task setting with a neural network potential (energy prediction), or adding the total energy of the respective molecule as an additional node feature (besides atom types), does not benefit or even decrease the model's performance throughout all datasets. This suggests that a physicochemical property, such as the total energy, can not be related to toxicity activity with the selected ML method. We also provide an attention weight analysis for helping to explain the black-box toxicity prediction of our model in 3D space and, thus, increase reliability in ML models. In summary, our findings highlight the relevance of considering 3D geometry information in EGNNs and provide a straightforward way to integrate molecular conformers into ML-based pipelines for toxicity prediction.

show abstract

The (Re)-Evolution of Quantitative Structure–Activity Relationship (QSAR) Studies Propelled by the Surge of Machine Learning Methods

Cited by 37 publications

References 47 publications

Biological Activity Predictions of Ligands Based on Hybrid Molecular Fingerprinting and Ensemble Learning

Biological Activity Predictions of Ligands Based on Hybrid Molecular Fingerprinting and Ensemble Learning

Guidance for good practice in the application of machine learning in development of toxicological quantitative structure-activity relationships (QSARs)

Equivariant Graph Neural Networks for Toxicity Prediction

Contact Info

Product

Resources

About