Ligand Biological Activity Predictions Using Fingerprint-Based Artificial Neural Networks (FANN-QSAR)

Myint, Kyaw Zeyar; Xie, Xiang‐Qun

doi:10.1007/978-1-4939-2239-0_9

Cited by 15 publications

(8 citation statements)

References 57 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In such cases, the interpretation of a structure-activity relationship (SAR) study is more practical from the descriptor perspective. However, regular features commonly used by the traditional ML models in current chemoinformatics studies to describe the small molecules, such as molecular fingerprints (21,63,82,83), physicochemical properties, topological properties, and thermodynamics properties (70), are not fully appropriate to be used in DL architecture (84). Thus, the development of more interpretable descriptors is dire.…”

Section: Discussion and Future Prospectivementioning

confidence: 99%

Deep Learning for Drug Design: an Artificial Intelligence Paradigm for Drug Discovery in the Big Data Era

Jing

Bian

et al. 2018

AAPS J

283

182

View full text Add to dashboard Cite

Over the last decade, deep learning (DL) methods have been extremely successful and widely used to develop artificial intelligence (AI) in almost every domain, especially after it achieved its proud record on computational Go. Compared to traditional machine learning (ML) algorithms, DL methods still have a long way to go to achieve recognition in small molecular drug discovery and development. And there is still lots of work to do for the popularization and application of DL for research purpose, e.g., for small molecule drug research and development. In this review, we mainly discussed several most powerful and mainstream architectures, including the convolutional neural network (CNN), recurrent neural network (RNN), and deep auto-encoder networks (DAENs), for supervised learning and nonsupervised learning; summarized most of the representative applications in small molecule drug design; and briefly introduced how DL methods were used in those applications. The discussion for the pros and cons of DL methods as well as the main challenges we need to tackle were also emphasized.

show abstract

Section: Discussion and Future Prospectivementioning

confidence: 99%

Deep Learning for Drug Design: an Artificial Intelligence Paradigm for Drug Discovery in the Big Data Era

Jing

Bian

et al. 2018

AAPS J

283

182

View full text Add to dashboard Cite

show abstract

“…In the present study, an in vitro receptor binding assay was performed using SPR and compared to the isotopic receptor binding assay, in order to establish a reliable and safe evaluation system to predict the psychoactivities of synthetic cannabinoids. There have been several previous attempts to create an in silico prediction model of synthetic cannabinoids (4,21,22). In order to establish an in silico prediction model, the collection of in vivo or in vitro data, such as the 50% maximal effective dose (ED 50 ) or pharmacodynamical values (K a , K d , K i ), is essential.…”

Section: Discussionmentioning

confidence: 99%

Receptor Binding Affinities of Synthetic Cannabinoids Determined by Non-Isotopic Receptor Binding Assay

et al. 2019

View full text Add to dashboard Cite

A major predictor of the efficacy of natural or synthetic cannabinoids is their binding affinity to the cannabinoid type I receptor (CB1) in the central nervous system, as the main psychological effects of cannabinoids are achieved via binding to this receptor. Conventionally, receptor binding assays have been performed using isotopes, which are inconvenient owing to the effects of radioactivity. In the present study, the binding affinities of five cannabinoids for purified CB1 were measured using a surface plasmon resonance (SPR) technique as a putative non-isotopic receptor binding assay. Results were compared with those of a radio-isotope-labeled receptor binding assay. The representative natural cannabinoid Δ9-tetrahydrocannabinol and four synthetic cannabinoids, JWH-015, JWH-210, RCS-4, and JWH-250, were assessed using both the SPR biosensor assay and the conventional isotopic receptor binding assay. The binding affinities of the test substances to CB1 were determined to be (from highest to lowest) 9.52 × 10−13 M (JWH-210), 6.54 × 10−12 M (JWH-250), 1.56 × 10−11 M (Δ9-tetrahydrocannabinol), 2.75 × 10−11 M (RCS-4), and 6.80 ×10−11 M (JWH-015) using the non-isotopic method. Using the conventional isotopic receptor binding assay, the same order of affinities was observed. In conclusion, our results support the use of kinetic analysis via SPR in place of the isotopic receptor binding assay. To replace the receptor binding affinity assay with SPR techniques in routine assays, further studies for method validation will be needed in the future.

show abstract

“…With molecular fingerprints ECFP6, FP2, MACCS combined with ANN models, the two-dimensional QSAR virtual screening can achieve an average r test value (which measures regression fitness) of 0.75 [ 114 , 115 ]. Deep learning multi-task neural networks worked so well that the AUC value for toxicity QSAR prediction of NIH/3T3 cells (mouse embryonic fibroblast) can reach 0.9, which is slightly higher than the AUC of 0.87 in random forests, in which molecular fingerprints as input of the model [ 116 ].…”

Section: Chemical Structure Based Toxicity Prediction By Machine Lmentioning

confidence: 99%

Machine Learning Based Toxicity Prediction: From Chemical Structural Description to Transcriptome Analysis

Wang

2018

IJMS

147

View full text Add to dashboard Cite

Toxicity prediction is very important to public health. Among its many applications, toxicity prediction is essential to reduce the cost and labor of a drug’s preclinical and clinical trials, because a lot of drug evaluations (cellular, animal, and clinical) can be spared due to the predicted toxicity. In the era of Big Data and artificial intelligence, toxicity prediction can benefit from machine learning, which has been widely used in many fields such as natural language processing, speech recognition, image recognition, computational chemistry, and bioinformatics, with excellent performance. In this article, we review machine learning methods that have been applied to toxicity prediction, including deep learning, random forests, k-nearest neighbors, and support vector machines. We also discuss the input parameter to the machine learning algorithm, especially its shift from chemical structural description only to that combined with human transcriptome data analysis, which can greatly enhance prediction accuracy.

show abstract

Ligand Biological Activity Predictions Using Fingerprint-Based Artificial Neural Networks (FANN-QSAR)

Cited by 15 publications

References 57 publications

Deep Learning for Drug Design: an Artificial Intelligence Paradigm for Drug Discovery in the Big Data Era

Deep Learning for Drug Design: an Artificial Intelligence Paradigm for Drug Discovery in the Big Data Era

Receptor Binding Affinities of Synthetic Cannabinoids Determined by Non-Isotopic Receptor Binding Assay

Machine Learning Based Toxicity Prediction: From Chemical Structural Description to Transcriptome Analysis

Contact Info

Product

Resources

About