PTML Model of ChEMBL Compounds Assays for Vitamin Derivatives
Determining the biological activity of vitamins derivatives is needed given that organic synthesis of analogs of vitamins is an active field of interest for Medicinal Chemistry, Pharmaceutical and Food Additives. Accordingly, scientists from different disciplines perform preclinical assays (nij) wit...
| Autores: | , , , , , |
|---|---|
| Tipo de recurso: | artículo |
| Fecha de publicación: | 2020 |
| País: | España |
| Institución: | Universidad del País Vasco |
| Repositorio: | Addi. Archivo Digital para la Docencia y la Investigación |
| OAI Identifier: | oai:addi.ehu.eus:10810/72302 |
| Acceso en línea: | http://hdl.handle.net/10810/72302 |
| Access Level: | acceso abierto |
| Palabra clave: | ChEMBL vitamins perturbation theory machinelearning big data multi-target models |
| Sumario: | Determining the biological activity of vitamins derivatives is needed given that organic synthesis of analogs of vitamins is an active field of interest for Medicinal Chemistry, Pharmaceutical and Food Additives. Accordingly, scientists from different disciplines perform preclinical assays (nij) with a considerable combination of assay conditions (cj). Indeed, ChEMBL platform contains a database that includes results from 36220 different biological activity bio-assays of 21240 different vitamin and vitamin derivatives. These assays present are heterogeneous in terms of assay combinations of cj. They are focused on > 500 different biological activity parameters (c0), > 340 different targets (c1), > 6200 types of cell (c2), > 120 organisms of assay (c3) and > 60 assay strains (c4). It includes a total of > 1850 niacin assays, > 1580 tretinoin assays, > 1580 retinol assays, 857 ascorbic acid assays, etc. Given the complexity of this combinatorial data in terms of being assimilated by researchers, we propose to build a model by combining Perturbation Theory (PT) basis and Machine Learning (ML). Through this study, we propose a PTML (PT + ML) combinatorial model for ChEMBL results on biological activity of vitamins and vitamins derivatives. The Linear Discriminant Analysis (LDA) model presented for training subset a Specificity (%) = 90.38, Sensitivity (%) = 87.51, and Accuracy (%) = 89.89. The model showed for external validation subset Specificity (%) = 90.58, Sensitivity (%) = 87.72, and Accuracy (%) = 90.09. Different types of linear and non-linear PTML models such as Logistic Regression (LR), Classification Tree (CT), Näive Bayes (NB), and Random Forest (RF) were applied in order to contrast the capacity of prediction. The PTML-LDA model predicts with more accuracy by applying combinatorial descriptors. In addition, PCA experiment with chemical structure descriptors allowed to characterize the high structural diversity of the chemical space studied. In any case, PTML models using chemical structure descriptors do not improve the performance of the PTML-LDA model based on LogP and PSA. We can conclude that the three variable PTML-LDA model is a simplified and adaptable tool for the prediction, for different experiment combinations, the biological activity of derivative vitamins. |
|---|