• JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
  • JoomlaWorks Simple Image Rotator
 
  Bookmark and Share
 
 
Master's Dissertation
DOI
https://doi.org/10.11606/D.9.2023.tde-10112023-174930
Document
Author
Full name
Raisa Ludmila Calil
E-mail
Institute/School/College
Knowledge Area
Date of Defense
Published
São Paulo, 2023
Supervisor
Committee
Trossini, Gustavo Henrique Goulart (President)
Alves, Vinicius de Medeiros
Honorio, Káthia Maria
Scotti, Marcus Tullius
Title in Portuguese
Uma análise quimioinformática sobre canabinoides sintéticos em receptores canabinóides (CB1 e CB2)
Keywords in Portuguese
Canabinóides sintéticos
HCA
MARS
PCA
Quimioinformática
Randon Forest
Abstract in Portuguese
Estudos científicos mostram o uso terapêutico da Cannabis sativa e de seus constituintes químicos, tais como: canabidiol, canabinol e tetrahidrocanabinol (THC), sendo este último responsável pelos efeitos psicoativos. Os receptores canabinoides CB1R e CB2R foram identificados, principalmente, no sistema nervoso central e imunológico, respectivamente. Canabinoides sintéticos são compostos que atuam nesses receptores, produzindo efeitos semelhantes à derivados canabinoides naturais. Técnicas computacionais são amplamente empregadas para avaliar seletividade molecular. Com o aumento das bibliotecas moleculares, se faz necessário explorar de forma racional as características estruturas e propriedades destes compostos. Neste estudo, utilizou-se técnicas de quimioinformática para gerar modelos que pudessem reconhecer e distinguir os perfis estruturais de canabinoides sintéticos e sua interação com os receptores canabinoides CB1R e CB2R. Para isso, foram realizadas: a construção de um banco de dados contendo informações sobre canabinoides sintéticos, cálculo de descritores moleculares, e construção de modelos utilizando técnicas como Análise Hierárquica de Agrupamentos (HCA), Análise de Componentes Principais (PCA), Árvores de decisão, Random Forest e Multivariate Adaptive Regression Splines (MARS). A seleção criteriosa dos descritores moleculares foi essencial para obter modelos estatísticos precisos. Quando se tratou de identificar as estruturas dos canabinoides sintéticos, o modelo baseado em árvore de decisão apresentou excelente desempenho na distinção da interação entre CB1R e CB2R. A redução da multicolinearidade mostrou que as abordagens de Forward Selection e Backward Elimination são eficazes na obtenção de modelos lineares simples e interpretáveis. O algoritmo Random Forest foi computacionalmente eficiente e proporcionou resultados confiáveis, enquanto o algoritmo MARS deve ser utilizado com cautela na predição de canabinoides sintéticos atuando em CB1R e CB2R. As técnicas de Análise do Componente Principal (PCA) e Análise Hierárquica de Agrupamentos (HCA) permitiram uma separação precisa dos dados e podem ser úteis em análises futuras.
Title in English
Uma análise quimioinformática sobre canabinoides sintéticos em receptores canabinóides (CB1 e CB2)
Keywords in English
Chemoinformatics
HCA
MARS
PCA
Random Forest
Synthetic cannabinoids
Abstract in English
Tetrahydrocannabinol (THC), the chemical component that gives cannabis its euphoric effects, is present in Cannabis sativa and has been shown in scientific research to be used therapeutically. CB1R and CB2R cannabinoid receptors have been found primarily in the central nervous and immune systems, respectively. Synthetic cannabinoids are substances that bind to these receptors and result in cannabis-like effects. With the increase of molecular libraries, computational tools are helpful for finding specific molecules, thus it is necessary to explore rationally. In this study, chemoinformatics techniques were used to create models that could recognize and distinguish the structural profiles of synthetic cannabinoids and study their interaction with the cannabinoid receptors CB1R and CB2R. ln order to achieve this, the following tasks were completed: building a database containing information on synthetic cannabinoids, estimating molecular descriptors, and constructing models using techniques like Hierarchical Cluster Analysis (HCA), Principal Component Analysis (PCA), Decision Trees, Random Forest and Multivariate Adaptive Regression Splines (MARS). To create correct statistical models, it was crucial to choose chemical descriptors carefully. When it carne to identifying the structures of synthetic cannabinoids that interact with CB1R and CB2R, the decision tree-based model performed exceptionally well. The decrease in multicollinearity demonstrated the efficacy of the Forward Selection and Backward Elimination procedures in producing straightforward and understandable linear models, while MARS algorithm should be used with caution in predicting synthetic cannabinoids acting on CB1R and CB2R. ln contrast, the Random Forest approach was computationally efficient and provided reliable results. The data may be separated precisely thanks to the Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA) approaches, which may be helpful in future research.
 
WARNING - Viewing this document is conditioned on your acceptance of the following terms of use:
This document is only for private use for research and teaching activities. Reproduction for commercial use is forbidden. This rights cover the whole data about this document as well as its contents. Any uses or copies of this document in whole or in part must include the author's name.
Publishing Date
2023-12-08
 
WARNING: Learn what derived works are clicking here.
All rights of the thesis/dissertation are from the authors
CeTI-SC/STI
Digital Library of Theses and Dissertations of USP. Copyright © 2001-2024. All rights reserved.