Developing novel computational prediction models for assessing chemical-induced neurotoxicity using naïve Bayes classifier technique

Zhang, Hui; Mao, Jun; Qi, Hua-Zhao; Xie, Hongtu; Shen, Chen; Liu, Chun‐Tao; Ding, Lei

doi:10.1016/j.fct.2020.111513

Cited by 15 publications

(8 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Structural alert (SA) was the toxophore (usually a specific substructure or fragment) which can lead to a particular toxicity endpoint. It has been widely used for toxicity research for many different toxicity endpoints (37)(38)(39)(40)(41)(42)(43)(44). In this study, the SAs for DIAD toxicity were detected by calculating the f-score (45) and frequency ratio of each fragment from KRFP fingerprint.…”

Section: Analysis Of Structural Alerts For Chemical Diad Toxicitymentioning

confidence: 99%

Modeling and insights into the structural characteristics of drug-induced autoimmune diseases

Guo

Zhang

et al. 2022

Front. Immunol.

View full text Add to dashboard Cite

The incidence and complexity of drug-induced autoimmune diseases (DIAD) have been on the rise in recent years, which may lead to serious or fatal consequences. Besides, many environmental and industrial chemicals can also cause DIAD. However, there are few effective approaches to estimate the DIAD potential of drugs and other chemicals currently, and the structural characteristics and mechanism of action of DIAD compounds have not been clarified. In this study, we developed the in silico models for chemical DIAD prediction and investigated the structural characteristics of DIAD chemicals based on the reliable drug data on human autoimmune diseases. We collected 148 medications which were reported can cause DIAD clinically and 450 medications that clearly do not cause DIAD. Several different machine learning algorithms and molecular fingerprints were combined to develop the in silico models. The best performed model provided the good overall accuracy on validation set with 76.26%. The model was made freely available on the website http://diad.sapredictor.cn/. To further investigate the differences in structural characteristics between DIAD chemicals and non-DIAD chemicals, several key physicochemical properties were analyzed. The results showed that AlogP, molecular polar surface area (MPSA), and the number of hydrogen bond donors (nHDon) were significantly different between the DIAD and non-DIAD structures. They may be related to the DIAD toxicity of chemicals. In addition, 14 structural alerts (SA) for DIAD toxicity were detected from predefined substructures. The SAs may be helpful to explain the mechanism of action of drug induced autoimmune disease, and can used to identify the chemicals with potential DIAD toxicity. The structural alerts have been integrated in a structural alert-based web server SApredictor (http://www.sapredictor.cn). We hope the results could provide useful information for the recognition of DIAD chemicals and the insights of structural characteristics for chemical DIAD toxicity.

show abstract

Section: Analysis Of Structural Alerts For Chemical Diad Toxicitymentioning

confidence: 99%

Modeling and insights into the structural characteristics of drug-induced autoimmune diseases

Guo

Zhang

et al. 2022

Front. Immunol.

View full text Add to dashboard Cite

show abstract

“…The negative association of descriptor [E_EDJ_equ] suggested that high molar volume and molar refractivity of a compound structure would reduce the inhibitory potency of a compound, and the further the position of heteroatom from the centre of the compound structure, the higher the inhibitory potency of the compound. Positive association of descriptor [Num_Rings] indicated a high number of base rings contribute to stronger inhibition of a compound towards β-tryptase [62]. The index value of descriptor [CIC] increases with a higher number of carbon atoms but decreases when the number of hydrogen atoms becomes smaller [63].…”

Section: D-qsar Modelmentioning

confidence: 99%

Insight parameter drug design for human beta-tryptase inhibition integrated molecular docking, QSAR, molecular dynamics simulation, and pharmacophore modelling studies of alpha-keto-[1,2,4]-oxadiazoles

Tan

Imran

et al. 2022

Preprint

View full text Add to dashboard Cite

Dengue hemorrhagic fever (DHF) is life-threatening severe dengue with a hallmark of vascular leakage. A mast cell protease, beta-tryptase, has been found to promote vascular leakage in DHF patients, which could be a potential target for the treatment of DHF. This study aims to develop a theoretical background for the design and selection of beta-tryptase inhibitors through these approaches: two-dimensional quantitative structure-activity relationships (2D-QSAR) study, molecular docking analysis, molecular dynamics (MD) simulation, and structure-based pharmacophore modelling (PM). A total of 34 beta-keto-[1,2,3]-oxadiazoles scaffold-based compounds, obtained from a past study, were used to generate 2D-QSAR models by Genetic Function Approximation (GFA). The generated 2D-QSAR models were used to investigate the relationships between the molecular structure and the potency of beta-tryptase inhibition. Molecular docking explores the binding affinities and binding interactions of the beta-keto-[1,2,3]-oxadiazoles scaffold-based compounds with beta-tryptase (PDB Code 4A6L) by the CDOCKER tool in Discovery Studio. In addition, MD simulation was performed using GROMACS on the docked complex of the reported most active compound, compound 11e, to study the binding mechanism of the compound towards beta-tryptase. Finally, a structure-based pharmacophore model was generated from the same docked complex to identify the important features that contribute positively to the inhibitory activity of the compound towards beta-tryptase. The best 2D-QSAR model has demonstrated statistically significant results through its r2, q2, and r2 (pred) values of 0.9077, 0.733, and 0.8104, respectively. The docking results of compound 11e showed lower CDOCKER energy than the 4A6L co-crystallised ligand, indicating good binding affinity. Furthermore, compound 11e has a similar binding pattern as the 4A6L co-crystallised ligand, which involves the binding of active residues such as Asp207, SER208, and GLY237. The MD simulation shows that the 4A6L-compound 11e complex has RMSD below 2 angstrom throughout the 100ns of simulation, indicating that the docked complex is stable. Besides, MD simulation showed that the inhibitory potency of compound 11e is contributed by hydrogen bonding with 4A6L active site residues, which are ASP207, SER208, and GLY237. The best pharmacophore model identified features that contribute to the inhibitory potency of a compound, which included hydrogen bond acceptor, ionic interaction, hydrophobic interaction, and aromatic ring. This study has fundamentally supplied valuable insight and knowledge for developing novel chemical compounds with improved inhibitory ability against human beta-tryptase.

show abstract

“…Research, Society andDevelopment, v. 10, n. 15, e140101522862, 2021 (CC BY 4.0) | ISSN 2525-3409 | DOI: http://dx.doi.org/10.33448/rsd-v10i15.22862 Uma vantagem deste algoritmo é saber lidar com vários dados de forma simultânea, com tolerância a ruídos aleatórios durante o processo de construção do modelo e foi aplicado para diferenciar neurotóxicos de não neurotóxicos, baseado em oito propriedades físico-químicas moleculares e dez tipos de impressões digitais moleculares com onze diâmetros diferentes. O Naive Bayes mostrou uma ferramenta robusta e desenvolveu com sucesso modelos de predição computacional induzidos por drogas de neurotoxicidade (Zhang et al, 2020).…”

Section: Naive Bayesunclassified

Aplicação de algoritmos de machine learning na área farmacêutica: uma revisão

Campos

Vasconcelos

2021

RSD

View full text Add to dashboard Cite

A indústria farmacêutica com toda sua importância vem inovando e revolucionando no decorrer do tempo. A tecnologia da informação e seus segmentos, tem um papel imprescindível para que as mudanças ocorram, e este projeto mostrará o cenário de crescimento da indústria farmacêutica e a importância de sua tecnologia no Brasil, no mundo e o uso de algoritmos como ferramentas essenciais em diversas áreas do campo farmacêutico. No decorrer deste projeto, até o seu fim, será apresentado pontos mostrando como está o cenário industrial, tecnologias farmacêuticas, algoritmos sendo imprescindíveis na resolução de problemas, alianças entre indústrias, inovações para novos medicamentos, atendimentos e tratamentos. Trata-se de uma revisão de literatura integrativa utilizando as plataformas Google Acadêmico, PubMed, Scielo e Science Direct para buscar de artigos do período de 2003 a 2021 sobre aplicação de algoritmos de machine learning na área farmacêutica. O uso de algoritmos se mostrou eficaz facilitando no desenvolvimento de novas drogas e para resolver problemas existentes.

show abstract

Developing novel computational prediction models for assessing chemical-induced neurotoxicity using naïve Bayes classifier technique

Cited by 15 publications

References 25 publications

Modeling and insights into the structural characteristics of drug-induced autoimmune diseases

Modeling and insights into the structural characteristics of drug-induced autoimmune diseases

Insight parameter drug design for human beta-tryptase inhibition integrated molecular docking, QSAR, molecular dynamics simulation, and pharmacophore modelling studies of alpha-keto-[1,2,4]-oxadiazoles

Aplicação de algoritmos de machine learning na área farmacêutica: uma revisão

Contact Info

Product

Resources

About