Kathia M. Honorio

Machine Learning Techniques and Drug Design

The interest in the application of machine learning techniques (MLT) as drug design tools is growing in the last decades. The reason for this is related to the fact that the drug design is very complex and requires the use of hybrid techniques. A brief review of some MLT such as self-organizing maps, multilayer perceptron, bayesian neural networks, counter-propagation neural network and support vector machines is described in this paper. A comparison between the performance of the described methods and some classical statistical methods (such as partial least squares and multiple linear regression) shows that MLT have significant advantages. Nowadays, the number of studies in medicinal chemistry that employ these techniques has considerably increased, in particular the use of support vector machines. The state of the art and the future trends of MLT applications encompass the use of these techniques to construct more reliable QSAR models. The models obtained from MLT can be used in virtual screening studies as well as filters to develop/discovery new chemicals. An important challenge in the drug design field is the prediction of pharmacokinetic and toxicity properties, which can avoid failures in the clinical phases. Therefore, this review provides a critical point of view on the main MLT and shows their potential ability as a valuable tool in drug design.

Electrochemical Behavior of Nicotine Studied by Voltammetric Techniques at Boron‐Doped Diamond Electrodes

Abstract The electrochemical behavior of nicotine in alkaline media was studied using a boron doped diamond (BDD) surface as the working electrode. In order to establish the pH dependence and to gain information about the mass transport of the species, cyclic voltammetry studies were carried out in a 0.1 mol L−1 BR (Britton‐Robinson) buffer in the presence of 1.0×10−3 mol L−1 nicotine. The optimum pH value was 8 and the mass transport was controlled by diffusion of the species. The square wave voltammetry technique was used to determine the electroanalytical parameters such as frequency, amplitude, and scan increment. After optimization, an analytical curve was constructed. The limits of detection and quantification were 0.50 and 1.66 mg L−1, respectively. Theoretical calculations indicate that the probable oxidation site on the nicotine molecule was the nitrogen atom denoted “11 N” and a speculation about the reaction mechanism was proposed. Finally, an experiment using a real sample (cigarette tobacco) was carried out and a recovery study revealed a value of about 4.3 mg L−1 in 0.1 g of tobacco.

Pharmacophore-based 3D QSAR studies on a series of high affinity 5-HT1A receptor ligands

5-HT(1A) receptor antagonists have been employed to treat depression, but the lack of structural information on this receptor hampers the design of specific and selective ligands. In this study, we have performed CoMFA studies on a training set of arylpiperazines (high affinity 5-HT(1A) receptor ligands) and to produce an effective alignment of the data set, a pharmacophore model was produced using Galahad. A statistically significant model was obtained, indicating a good internal consistency and predictive ability for untested compounds. The information gathered from our receptor-independent pharmacophore hypothesis is in good agreement with results from independent studies using different approaches. Therefore, this work provides important insights on the chemical and structural basis involved in the molecular recognition of these compounds.

Applying machine learning techniques for ADME-Tox prediction: a review

Introduction: Pharmacokinetics involves the study of absorption, distribution, metabolism, excretion and toxicity of xenobiotics (ADME-Tox). In this sense, the ADME-Tox profile of a bioactive compound can impact its efficacy and safety. Moreover, efficacy and safety were considered some of the major causes of clinical failures in the development of new chemical entities. In this context, machine learning (ML) techniques have been often used in ADME-Tox studies due to the existence of compounds with known pharmacokinetic properties available for generating predictive models. Areas covered: This review examines the growth in the use of some ML techniques in ADME-Tox studies, in particular supervised and unsupervised techniques. Also, some critical points (e.g., size of the data set and type of output variable) must be considered during the generation of models that relate ADME-Tox properties and biological activity. Expert opinion: ML techniques have been successfully employed in pharmacokinetic studies, helping the complex process of designing new drug candidates from the use of reliable ML models. An application of this procedure would be the prediction of ADME-Tox properties from studies of quantitative structure–activity relationships or the discovery of new compounds from a virtual screening using filters based on results obtained from ML techniques.

Aspectos terapêuticos de compostos da planta Cannabis sativa

Several cannabinoid compounds present therapeutic properties, but also have psychotropic effects, limiting their use as medicine. Nowadays, many important discoveries on the compounds extracted from the plant Cannabis sativa (cannabinoids) have contributed to understand the therapeutic properties of these compounds. The main discoveries in the last years on the cannabinoid compounds were: the cannabinoid receptors CB1 and CB2, the endogenous cannabinoids and the possible mechanisms of action involved in the interaction between cannabinoid compounds and the biological receptors. So, from the therapeutical aspects presented in this work, we intended to show the evolution of the Cannabis sativa research and the possible medicinal use of cannabinoid compounds.

A structure–activity relationship (SAR) study of synthetic neolignans and related compounds with biological activity against Escherichia coli

Abstract Structure–activity relationship techniques were employed to classify neolignan compounds and related analogues against Escherichia coli . The AM1 (Austin Model) method was used to calculate a set of molecular descriptors (properties) for 16 synthetic neolignan compounds. The descriptors were further analyzed using the principal component analysis (PCA), hierarchical cluster analysis (HCA) and K-nearest neighbor (KNN) chemometric methods. The PCA and HCA methods showed quite efficient to classify the 16 compounds in two groups (active and inactive) and three descriptors were found to be important in the classification. By using the chemometric results, 14 new neolignan molecules were analyzed through the PCA, HCA and KNN methods and 11 of them are proposed as molecules potentially active against E. coli .

Propriedades químico-quânticas empregadas em estudos das relações estrutura-atividade

In this work we show that structure-activity relationship studies are of great importance in modern chemistry and biochemistry. In order to obtain a significant correlation, it is crucial that appropriate descriptors be employed. Thus, quantum chemical calculations are an attractive source of new molecular descriptors which can, in principle, express all the electronic and geometric properties of molecules and their interactions with the biological receptor.

A Structure-Activity Relationship (SAR) Study of Neolignan Compounds with Anti-schistosomiasis Activity

A set of eighteen neolignan derivative compounds with anti-schistosomiasis activity was studied by using the quantum mechanical semi-empirical method PM3 and other theoretical methods in order to calculate selected molecular properties (variables or descriptors) to be correlated to their biological activities. Exploratory data analysis (principal component analysis, PCA, and hierarchical cluster analysis, HCA), discriminant analysis (DA) and the Kth nearest neighbor (KNN) method were employed for obtaining possible relationships between the calculated descriptors and the biological activities studied and predicting the anti-schistosomiasis activity of new compounds from a test set. The molecular descriptors responsible for the separation between active and inactive compounds were: hydration energy (HE), molecular refractivity (MR) and charge on the C19 carbon atom (Q19). These descriptors give information on the kind of interaction that can occur between the compounds and their respective biological receptor. The prediction study was done with a new set of ten derivative compounds by using the PCA, HCA, DA and KNN methods and only five of them were predicted as active against schistosomiasis.

Estudo eletroquímico e químico-quântico da oxidação do antidepressivo tricíclico amitriptilina

This work presents the electrochemical and quantum chemical studies of the oxidation of the tricyclic antidepressant amitriptyline (AM) employing a carbon-polyurethane composite electrode (GPU) in a 0.1 mol L-1 BR buffer. The electrochemical results showed that the oxidation of AM occurs irreversibly at potentials close to 830 mV with the loss of one electron and one proton and is controlled by reagent and product adsorption. According to the PM3 results, the atom C16 is the region of highest probability for the oxidation of AM since it has the largest charge variation.

Advanced QSAR Studies on PPARd Ligands Related to Metabolic Diseases

PPARd e um receptor nuclear que, quando ativado, regula o metabolismo de carboidratos e lipidios, e esta relacionado com diversas enfermidades, tais como sindrome metabolica e diabetes tipo 2. Para entender as principais interacoes entre alguns ligantes bioativos e o receptor PPARd, modelos de QSAR 2D e 3D foram obtidos e comparados com mapas de potencial eletrostatico (MEP) e dos orbitais de fronteira (HOMO e LUMO), assim como resultados de docagem molecular. Os modelos de QSAR obtidos apresentaram bons resultados estatisticos e foram utilizados para predizer a atividade biologica de compostos do conjunto-teste (validacao externa), e os valores preditos estao em concordância com os resultados experimentais. Alem disso, todos mapas moleculares foram utilizados para avaliar as possiveis interacoes entre os ligantes e o receptor PPARd. Portanto, os modelos de QSAR 2D e 3D, assim como os mapas de HOMO, LUMO e MEP, podem fornecer informacoes sobre as principais propriedades necessarias para o planejamento de novos ligantes do receptor PPARd. PPARd is a nuclear receptor that, when activated, regulates the metabolism of carbohydrates and lipids and is related to metabolic syndrome and type 2 diabetes. To understand the main interactions between ligands and PPARd, we have constructed 2D and 3D QSAR models and compared them with HOMO, LUMO and electrostatic potential maps of the compounds studied, as well as docking results. All QSAR models showed good statistical parameters and prediction outcomes. The QSAR models were used to predict the biological activity of an external test set, and the predicted values are in good agreement with the experimental results. Furthermore, we employed all maps to evaluate the possible interactions between the ligands and PPARd. These predictive QSAR models, along with the HOMO, LUMO and MEP maps, can provide insights into the structural and chemical properties that are needed in the design of new PPARd ligands that have improved biological activity and can be employed to treat metabolic diseases.