Simplified molecular input line entry system-based descriptors in QSAR modeling for HIV-protease inhibitors

Islam, Ataul; Pillay, Tahir S

doi:10.1016/j.chemolab.2016.02.008

Cited by 37 publications

(9 citation statements)

References 52 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The CORAL software (http://www.insilico.eu/coral) is a tool to build up QSAR models for endpoints related to medicinal chemistry (Veselinović et al, 2013;Worachartcheewan et al, 2014;Li et al, 2014;Ghaedi, 2015;Nesmerak et al, 2015;Islam and Pillay, 2016). In addition, the classification models for anti-sarcoma activity (Toropov et al, 2012a) and for liver-related adverse effects of drugs (Toropov et al, 2012b) were suggested.…”

Section: Introductionmentioning

confidence: 99%

CORAL: Binary classifications (active/inactive) for drug-induced liver injury

Toropova

Toropov

2017

Toxicology Letters

View full text Add to dashboard Cite

The approach has been checked up with a group of random splits into the training and validation sets. These stochastic experiments have shown the stability of results: predictability of the models for various splits. Thus, the attempt to build up the classification QSAR model by means of the Monte Carlo technique, based on representation of the molecular structure via simplified molecular input line entry systems (SMILES) and hydrogen suppressed graph (HSG) using the CORAL software (http://www.insilico.eu/coral) has shown ability of this approach to provide quite good prediction of the examined endpoint (drug-induced liver injury).

show abstract

Section: Introductionmentioning

confidence: 99%

CORAL: Binary classifications (active/inactive) for drug-induced liver injury

Toropova

Toropov

2017

Toxicology Letters

View full text Add to dashboard Cite

show abstract

“…The C5 and C6 are descriptors characterizing rings in the molecular structure. These descriptors are calculated with the molecular graph (23)(24)(25). C5 and C6 are codes sensitive to the number of corresponding rings in the molecular structure, the presence (absence) heteroatoms, and the presence (absence) of aromaticity.…”

Section: Methodsmentioning

confidence: 99%

“…The aim of the present study was to improve the CORAL model described in (1) by means of using two approaches, namely (i) using of the Index of Ideality of Correlation (IIC), which is a new criterion for the predictive potential of QSAR models (19)(20)(21)(22); and (ii) using correlation weights, which are related to the presence of different rings in the molecular structure (23)(24)(25).…”

mentioning

confidence: 99%

Predicting Cytotoxicity of 2-Phenylindole Derivatives Against Breast Cancer Cells Using Index of Ideality of Correlation

Toropov

Toropova

2018

Anticancer Res

View full text Add to dashboard Cite

Background: Breast cancer is one of the leading types of cancer in women worldwide. Quantitative structureactivity relationship (QSAR) methods play an important role in the search for new anticancer agents. A QSAR model for cytotoxicity against the breast cancer cell line MCF7, based on hybrid optimal descriptors, has been suggested. A modified version of the hybrid descriptor is suggested. Materials and Methods: A QSAR model for the anticancer activity of 2-phenylindole derivatives was built using the Index of Ideality of Correlation (IIC), which is a new criterion for predictive potential. The calculation can be carried out with a modified version of the CORAL software. Results: The model for the anticancer activity suggested here is better than the one described in the literature. Conclusion: Taking into account the data on molecular rings together with the use of new criterion of predictive potential (IIC), the QSAR improves the prediction for anticancer activity. Anticancer drug discovery is a complex and important field of natural sciences. Quantitative structure-activity relationships (QSARs) are not able to provide complete data on molecular architecture required for new anticancer agents, but QSARs can help reduce the time needed and cost of the search for such agents (1). The design of new chemical compounds that are active against the breast cancer cell line MCF7 has several conceptually different approaches. These are the well-known ADMET approach (absorption, distribution, metabolism, excretion, and toxicity) (2); general virtual screening based on 6189 This article is freely accessible online.

show abstract

“…The existence of important drug experimental trials databases such as ChEMBL has provided the opportunity to search for new relevant information, due to its capacity to deliver huge amounts of data . The application of cheminformatics methodology and other computational methods increases drug design and best candidate selection for experimental trials, providing the relationship insights into structural molecular information contributing to biological activity …”

Section: Introductionmentioning

confidence: 99%

Multioutput Perturbation-Theory Machine Learning (PTML) Model of ChEMBL Data for Antiretroviral Compounds

et al. 2019

View full text Add to dashboard Cite

Retroviral infections, such as HIV, are, until now, diseases with no cure. Medicine and pharmaceutical chemistry need and consider it a huge goal to define target proteins of new antiretroviral compounds. ChEMBL manages Big Data features with a complex data set, which is hard to organize. This makes information difficult to analyze due to a big number of characteristics described in order to predict new drug candidates for retroviral infections. For this reason, we propose to develop a new predictive model combining perturbation theory (PT) bases and machine learning (ML) modeling to create a new tool that can take advantage of all the available information. The PTML model proposed in this work for the ChEMBL data set preclinical experimental assays for antiretroviral compounds consists of a linear equation with four variables. The PT operators used are founded on multicondition moving averages, combining different features and simplifying the difficulty to manage all data. More than 140 000 preclinical assays for 56 105 compounds with different characteristics or experimental conditions have been carried out and can be found in ChEMBL database, covering combinations with 359 biological activity parameters (c 0 ), 55 protein accessions (c 1 ), 83 cell lines (c 2 ), 64 organisms of assay (c 3 ), and 773 subtypes or strains. We have included 150 148 preclinical experimental assays for HIV virus, 1188 for HTLV virus, 84 for simian immunodeficiency virus, 370 for murine leukemia virus, 119 for Rous sarcoma virus, 1581 for MMTV, etc. We also included 5277 assays for hepatitis B virus. The developed PTML model reached considerable values in sensibility (73.05% for training and 73.10% for validation), specificity (86.61% for training and 87.17% for validation), and accuracy (75.84% for training and 75.98% for validation). We also compared alternative PTML models with different PT operators such as covariance, moments, and exponential terms. Finally, we made a comparison between literature ML models with our PTML model and also artificial neural network (ANN) nonlinear models. We conclude that this PTML model is the first one to consider multiple characteristics of preclinical experimental antiretroviral assays combined, generating a simple, useful, and adaptable instrument, which could reduce time and costs in antiretroviral drugs research.

show abstract

Simplified molecular input line entry system-based descriptors in QSAR modeling for HIV-protease inhibitors

Cited by 37 publications

References 52 publications

CORAL: Binary classifications (active/inactive) for drug-induced liver injury

CORAL: Binary classifications (active/inactive) for drug-induced liver injury

Predicting Cytotoxicity of 2-Phenylindole Derivatives Against Breast Cancer Cells Using Index of Ideality of Correlation

Multioutput Perturbation-Theory Machine Learning (PTML) Model of ChEMBL Data for Antiretroviral Compounds

Contact Info

Product

Resources

About