Drug Design with Artificial Intelligence Methods

Ivanciuc, Ovidiu

doi:10.1007/978-0-387-30440-3_133

Cited by 6 publications

(5 citation statements)

References 100 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Several researchers have demonstrated the high performance of different types of computational classifiers in structure−function relationship studies ranging from low-weight molecules to protein or protein−protein complexes and based on different algorithms; see for instance the works of Ivanciuc about Machine Learning, − or the works of Cai and Chou et al − ,− with different classifiers. In particular, the Linear Neural Network (LNN) algorithm, the simpler type of ANN, was used here to train different linear models based on different combinations of parameters.…”

Section: Resultsmentioning

confidence: 99%

Trypano-PPI: A Web Server for Prediction of Unique Targets in Trypanosome Proteome by using Electrostatic Parameters of Protein−protein Interactions

et al. 2009

View full text Add to dashboard Cite

Trypanosoma brucei causes African trypanosomiasis in humans (HAT or African sleeping sickness) and Nagana in cattle. The disease threatens over 60 million people and uncounted numbers of cattle in 36 countries of sub-Saharan Africa and has a devastating impact on human health and the economy. On the other hand, Trypanosoma cruzi is responsible in South America for Chagas disease, which can cause acute illness and death, especially in young children. In this context, the discovery of novel drug targets in Trypanosome proteome is a major focus for the scientific community. Recently, many researchers have spent important efforts on the study of protein-protein interactions (PPIs) in pathogen Trypanosome species concluding that the low sequence identities between some parasite proteins and their human host render these PPIs as highly promising drug targets. To the best of our knowledge, there are no general models to predict Unique PPIs in Trypanosome (TPPIs). On the other hand, the 3D structure of an increasing number of Trypanosome proteins is reported in databases. In this regard, the introduction of a new model to predict TPPIs from the 3D structure of proteins involved in PPI is very important. For this purpose, we introduced new protein-protein complex invariants based on the Markov average electrostatic potential xi(k)(R(i)) for amino acids located in different regions (R(i)) of i-th protein and placed at a distance k one from each other. We calculated more than 30 different types of parameters for 7866 pairs of proteins (1023 TPPIs and 6823 non-TPPIs) from more than 20 organisms, including parasites and human or cattle hosts. We found a very simple linear model that predicts above 90% of TPPIs and non-TPPIs both in training and independent test subsets using only two parameters. The parameters were (d)xi(k)(s) = |xi(k)(s(1)) - xi(k)(s(2))|, the absolute difference between the xi(k)(s(i)) values on the surface of the two proteins of the pairs. We also tested nonlinear ANN models for comparison purposes but the linear model gives the best results. We implemented this predictor in the web server named TrypanoPPI freely available to public at http://miaja.tic.udc.es/Bio-AIMS/TrypanoPPI.php. This is the first model that predicts how unique a protein-protein complex in Trypanosome proteome is with respect to other parasites and hosts, opening new opportunities for antitrypanosome drug target discovery.

show abstract

Section: Resultsmentioning

confidence: 99%

Trypano-PPI: A Web Server for Prediction of Unique Targets in Trypanosome Proteome by using Electrostatic Parameters of Protein−protein Interactions

et al. 2009

View full text Add to dashboard Cite

show abstract

“…To overcome this difficulty, feature selection techniques are employed to select those features relevant to describe the observed structure–activity relationships. It has also been shown that machine learning optimization algorithms such as Genetic Algorithms (GA), Particle Swarm Optimization (PSO), Artificial Immune System (AIS), and Ant Colony Optimization (ACO) are very powerful in QSAR modeling for feature selection. − The wide range of application of machine learning techniques to QSAR modeling has been extensively reviewed elsewhere. − …”

Section: Introductionmentioning

confidence: 99%

GA(M)E-QSAR: A Novel, Fully Automatic Genetic-Algorithm-(Meta)-Ensembles Approach for Binary Classification in Ligand-Based Drug Design

Pérez-Castillo

Lazar

Taminau

et al. 2012

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Computer-aided drug design has become an important component of the drug discovery process. Despite the advances in this field, there is not a unique modeling approach that can be successfully applied to solve the whole range of problems faced during QSAR modeling. Feature selection and ensemble modeling are active areas of research in ligand-based drug design. Here we introduce the GA(M)E-QSAR algorithm that combines the search and optimization capabilities of Genetic Algorithms with the simplicity of the Adaboost ensemble-based classification algorithm to solve binary classification problems. We also explore the usefulness of Meta-Ensembles trained with Adaboost and Voting schemes to further improve the accuracy, generalization, and robustness of the optimal Adaboost Single Ensemble derived from the Genetic Algorithm optimization. We evaluated the performance of our algorithm using five data sets from the literature and found that it is capable of yielding similar or better classification results to what has been reported for these data sets with a higher enrichment of active compounds relative to the whole actives subset when only the most active chemicals are considered. More important, we compared our methodology with state of the art feature selection and classification approaches and found that it can provide highly accurate, robust, and generalizable models. In the case of the Adaboost Ensembles derived from the Genetic Algorithm search, the final models are quite simple since they consist of a weighted sum of the output of single feature classifiers. Furthermore, the Adaboost scores can be used as ranking criterion to prioritize chemicals for synthesis and biological evaluation after virtual screening experiments.

show abstract

“…The drug design and development process involves the use of a variety of computational techniques, such as (quantitative) structure-activity relationships [(Q)SAR], molecular mechanics, quantum mechanics, molecular dynamics, and drug-receptor docking [ 1 , 2 ]. (Q)SAR studies are based on the premise that biological response is a function of the chemical structure [ 3 , 4 ].…”

Section: Introductionmentioning

confidence: 99%

Diverse models for anti-HIV activity of purine nucleoside analogs

Khatri

Lather²,

Мадан

2015

Chemistry Central Journal

View full text Add to dashboard Cite

BackgroundPurine nucleoside analogs (PNAs) constitute an important group of cytotoxic drugs for the treatment of neoplastic and autoimmune diseases. In the present study, classification models have been developed for the prediction of the anti-HIV activity of purine nucleoside analogs.ResultsThe topochemical version of superaugmented pendentic index-4 has been proposed and successfully utilized for the development of models. A total of 60 2D and 3D molecular descriptors (MDs) of diverse nature were selected for building the classification models using decision tree (DT), random forest (RF), support vector machine (SVM), and moving average analysis (MAA). The values of most of these descriptors for each of the analogs in the dataset were computed using the Dragon software (version 5.3). An in-house computer program was also employed to calculate additional MDs which were not included in the Dragon software. DT, RF, and SVM correctly classified the analogs into actives and inactives with an accuracy of 89 %, 83 %, and 78 %, respectively. MAA-based models predicted the anti-HIV activity of purine nucleoside analogs with a non-error rate up to 98 %. Therapeutic active spans of the suggested MAA-based models not only showed more potency but also exhibited enhanced safety as revealed by comparatively high values of selectivity index (SI). The statistical importance of the developed models was appraised via intercorrelation analysis, specificity, sensitivity, non-error rate, and Matthews correlation coefficient.ConclusionsHigh predictability of the proposed models clearly indicates an immense potential for developing lead molecules for potent but safe anti-HIV purine nucleoside analogs.

show abstract

Drug Design with Artificial Intelligence Methods

Cited by 6 publications

References 100 publications

Trypano-PPI: A Web Server for Prediction of Unique Targets in Trypanosome Proteome by using Electrostatic Parameters of Protein−protein Interactions

Trypano-PPI: A Web Server for Prediction of Unique Targets in Trypanosome Proteome by using Electrostatic Parameters of Protein−protein Interactions

GA(M)E-QSAR: A Novel, Fully Automatic Genetic-Algorithm-(Meta)-Ensembles Approach for Binary Classification in Ligand-Based Drug Design

Diverse models for anti-HIV activity of purine nucleoside analogs

Contact Info

Product

Resources

About