Deep Neural Networks for QSAR

Xu, Yuting

doi:10.1007/978-1-0716-1787-8_10

Cited by 24 publications

(17 citation statements)

References 101 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, there has been significant progress in applying machine learning methods to QSAR predictions. − For example, an average MAE of ∼36 mV was achieved for predicting the experimentally determined midpoint redox potentials of flavoprotein enzymes using quantitative structure–property relationships based on extreme gradient boosting machine learning . However, studies describing deep learning of the redox potentials of organic molecules are both very limited and very recent.…”

Section: Deep Learningmentioning

confidence: 99%

Unlocking the Potential: Predicting Redox Behavior of Organic Molecules, from Linear Fits to Neural Networks

Fedorov

Gryn’ova

2023

J. Chem. Theory Comput.

View full text Add to dashboard Cite

Redox-active organic molecules, i.e., molecules that can relatively easily accept and/or donate electrons, are ubiquitous in biology, chemical synthesis, and electronic and spintronic devices, such as solar cells and rechargeable batteries, etc. Choosing the best candidates from an essentially infinite chemical space for experimental testing in a target application requires efficient screening approaches. In this Review, we discuss modern in silico techniques for predicting reduction and oxidation potentials of organic molecules that go beyond conventional first-principles computations and thermodynamic cycles. Approaches ranging from simple linear fits based on molecular orbital energy approximation and energy difference approximation to advanced regression and neural network machine learning algorithms employing complex descriptors of molecular compositions, geometries, and electronic structures are examined in conjunction with relevant literature examples. We discuss the interplay between ab initio data and machine learning (ML), i.e., whether it is better to base predictions on low-level quantum-chemical results corrected with ML or to bypass first-principles computations entirely and instead rely on elaborate deep learning architectures. Finally, we list currently available data sets of redox-active organic molecules and their experimental and/or computed properties to facilitate the development of screening platforms and rational design of redoxactive organic molecules.

show abstract

Section: Deep Learningmentioning

confidence: 99%

Unlocking the Potential: Predicting Redox Behavior of Organic Molecules, from Linear Fits to Neural Networks

Fedorov

Gryn’ova

2023

J. Chem. Theory Comput.

View full text Add to dashboard Cite

show abstract

“…These models are typically used to rank a list of candidate molecules for future laboratory experiments and to assist chemists in better understanding how structural changes affect a molecule's biological activities. 25 4) ORGANIC: It is an efficient molecular generation tool to create molecules with desired properties. It can be utilized by accessing the URL: https://github.…”

Section: Employing Ai In Drug Discoverymentioning

confidence: 99%

Impact of Artificial Intelligence (AI) on Drug Discovery and Product Development

Narayanan¹,

Durga²,

Nagalakshmi³

2022

Ind. J. Pharm. Edu. Res

View full text Add to dashboard Cite

Artificial Intelligence (AI) had transfigured different sectors in society, where the pharmaceutical sector is not an exceptional case. Pharmaceutical sectors have reached new heights with the emergence of these sophisticated technologies. The evolution of artificial intelligence in the pharmaceutical industry is in a growth phase opening the possibilities of discovering many new drugs. The diseases affecting humans are increasing tremendously whereas the drugs which are available to treat or cure are very much minimal. But this kind of scenario will not be present in the future because of the combination of artificial intelligence and the pharmaceutical industry which results in faster discovery of drugs with increased clinical outcomes. There was a shift in the paradigm of various stages in drug discovery because of the utilization of artificial intelligence. Each stage of drug discovery involves a certain timeline that can be cut down with the help of artificial intelligence. Many pharma companies are engaged with AI-based drug discovery approaches for treating various diseases like Parkinson's disease diabetes, Alzheimer's, Obsessive Compulsive Disorder, etc., AI is also being employed in product development for the fabrication of nanomedicines and nanorobots. Few AI-based drugs are already in the phase of clinical trials which indicates the growth of AI-driven drug discovery. In this review, we have highlighted the application of AI in drug discovery and product development of pharmaceuticals.

show abstract

“…10,11 Deep learning is a machine learning method which specifically employs the deep neural network algorithm to transform input molecular features into an activity prediction output through interconnected layers of neuron units comprised of numerical weights and biases. 12 Neural networks are trained to learn QSAR patterns using backpropagation where these weight and bias values are iteratively optimized to minimize a loss function quantifying the difference between the predicted and actual activities, until an ideal set of neurons is obtained which accurately transform molecular feature inputs into the correct activity prediction. 12 Neural networks can be developed to predict each S. typhimurium ± S9 activity label in a process defined as "single task" learning; i.e., each neural network learns and outputs a single predictive task (Figure 1…”

Section: ■ Introductionmentioning

confidence: 99%

Mechanistic Task Groupings Enhance Multitask Deep Learning of Strain-Specific Ames Mutagenicity

Lui

Guan

Matthews

2023

Chem. Res. Toxicol.

View full text Add to dashboard Cite

The Ames test is a gold standard mutagenicity assay that utilizes various Salmonella typhimurium strains with and without S9 fraction to provide insights into the mechanisms by which a chemical can mutate DNA. Multitask deep learning is an ideal framework for developing QSAR models with multiple end points, such as the Ames test, as the joint training of multiple predictive tasks may synergistically improve the prediction accuracy of each task. This work investigated how toxicology domain knowledge can be used to handcraft task groupings that better guide the training of multitask neural networks compared to a naïve ungrouped multitask neural network developed on a complete set of tasks. Sixteen S. typhimurium ± S9 strain tasks were used to generate groupings based on mutagenic and metabolic mechanisms that were reflected in correlation data analyses. Both grouped and ungrouped multitask neural networks predicted the 16 strain tasks with a higher balanced accuracy compared with single task controls, with grouped multitask neural networks consistently featuring incremental increases in predictivity over the ungrouped approach. We conclude that the main variable driving these performance improvements is the general multitask effect with mechanistic task groupings acting as an enhancement step to further concentrate synergistic training signals united by a common biological mechanism. This approach enables incorporation of toxicology domain knowledge into multitask QSAR model development allowing for more transparent and accurate Ames mutagenicity prediction.

show abstract

Deep Neural Networks for QSAR

Cited by 24 publications

References 101 publications

Unlocking the Potential: Predicting Redox Behavior of Organic Molecules, from Linear Fits to Neural Networks

Unlocking the Potential: Predicting Redox Behavior of Organic Molecules, from Linear Fits to Neural Networks

Impact of Artificial Intelligence (AI) on Drug Discovery and Product Development

Mechanistic Task Groupings Enhance Multitask Deep Learning of Strain-Specific Ames Mutagenicity

Contact Info

Product

Resources

About