Could deep learning in neural networks improve the QSAR models?

Gini, Giuseppina; Zanoli, F.; Gamba, Alessio; Raitano, Giuseppa; Benfenati, Emilio

doi:10.1080/1062936x.2019.1650827

Cited by 31 publications

(18 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Besides these features, there are many research works that use more raw representations of a molecule. A SMILES string representation of a molecule can be used as input for 1D convolutional [16,[40][41][42] or recurrent neural network [42][43]. This approach performs well in chromatography-related tasks [16,44].…”

Section: Introductionmentioning

confidence: 99%

Gas Chromatographic Retention Index Prediction Using Multimodal Machine Learning

Matyushin

Buryak

2020

IEEE Access

View full text Add to dashboard Cite

Gas chromatography is a widely used method in analytical chemistry and metabolomics. Using gas chromatography, vaporizable compounds can be separated for their further identification. Retention indices are standardized values that depend only on a chemical structure of a compound and on a stationary phase and characterize the retention of a compound in a chromatographic system. Retention index prediction is an important task because databases contain experimental values for a small fraction of all possible molecules, while this information is usable for untargeted analysis. In this work, we consider four machine learning models for retention index prediction: 1D and 2D convolutional neural networks, deep residual multilayer perceptron, and gradient boosting. String representation of the molecule, 2D representation of the chemical structure, molecular descriptors and fingerprints, and molecular descriptors are used as inputs of these four models, respectively, along with information about the stationary phase. The first and third models show the best performance, while the other two perform slightly worse. The models predict retention index values for various standard and semi-standard non-polar stationary phases. Further improvement in performance was achieved using a linear model that uses the results of four previous models as inputs (model stacking). The models were tested using various diverse data sets: flavor compounds, essential oils, metabolomics-related compounds. Achieved accuracy: median absolute and percentage errors -6-40 units and 0.8-2.2%. Accuracy depends on a test data set. The stacking model outperforms previously reported approaches for all test data sets. Parameters of a pre-trained model and some source code are provided.INDEX TERMS Analytical chemistry, convolutional neural network, deep learning, gas chromatography, gradient boosting, residual neural network, retention index, untargeted chemical analysis.

show abstract

Section: Introductionmentioning

confidence: 99%

Gas Chromatographic Retention Index Prediction Using Multimodal Machine Learning

Matyushin

Buryak

2020

IEEE Access

View full text Add to dashboard Cite

show abstract

“…12,13 DL has also been applied in chemistry, such as informing high throughput screening (HTS), Quantitative Structure Activity Relationship (QSAR) analyses, and others. [14][15][16] One important application of biological and chemical DL is in understanding natural product diversity and chemistry. This is especially evident in the applications of DL natural language processing (NLP) methods.…”

Section: Advances In Deep Learningmentioning

confidence: 99%

The application potential of machine learning and genomics for understanding natural product diversity, chemistry, and therapeutic translatability

Prihoda

Maritz

Klempíř³

et al. 2021

Nat. Prod. Rep.

View full text Add to dashboard Cite

show abstract

“…Apart from images, SMILES can also be used to predict the toxicity of a molecule, as implemented in SmilesNet developed by Gini and coworkers. It uses SMILES strings as an input, transforms them into a feature vector, and subsequently predicts the mutagenicity of a compound . A representation which seems to be the most natural in light of representing molecules as atoms and bonds is the molecule representation by graphs.…”

Section: Machine Learning Based Predictionsmentioning

confidence: 99%

“…Although all methods are very promising, the problem of the interpretability still exists, especially for multitask networks. However, interesting conclusions were drawn by Mayr et al, Gini et al, and Xu and coworkers who could show that networks can learn representations which are comparable to structural alerts . Wenzel and coworkers introduced response maps to highlight important features used by the network …”

Section: Machine Learning Based Predictionsmentioning

confidence: 99%

In silico toxicology: From structure–activity relationships towards deep learning and adverse outcome pathways

Hemmerich

Ecker

2020

WIREs Comput Mol Sci

110

View full text Add to dashboard Cite

In silico toxicology is an emerging field. It gains increasing importance as research is aiming to decrease the use of animal experiments as suggested in the 3R principles by Russell and Burch. In silico toxicology is a means to identify hazards of compounds before synthesis, and thus in very early stages of drug development. For chemical industries, as well as regulatory agencies it can aid in gap‐filling and guide risk minimization strategies. Techniques such as structural alerts, read‐across, quantitative structure–activity relationship, machine learning, and deep learning allow to use in silico toxicology in many cases, some even when data is scarce. Especially the concept of adverse outcome pathways puts all techniques into a broader context and can elucidate predictions by mechanistic insights. This article is categorized under: Structure and Mechanism > Computational Biochemistry and Biophysics Data Science > Chemoinformatics

show abstract

Could deep learning in neural networks improve the QSAR models?

Cited by 31 publications

References 20 publications

Gas Chromatographic Retention Index Prediction Using Multimodal Machine Learning

Gas Chromatographic Retention Index Prediction Using Multimodal Machine Learning

The application potential of machine learning and genomics for understanding natural product diversity, chemistry, and therapeutic translatability

In silico toxicology: From structure–activity relationships towards deep learning and adverse outcome pathways

Contact Info

Product

Resources

About