MIA‐QSAR evaluation of a series of sulfonylurea herbicides

Abstract

Both MIA-QSAR models showed high predictive ability, comparable with that of a reference methodology based on 3D descriptors. The method is suggested as a suitable tool for predicting novel herbicides.

Multi-Objective Modeling of Herbicidal Activity from an Environmentally Friendly Perspective

“…In this sense, the multivariate image analysis applied to quantitative structure-property relationships (MIA-QSPR) method appears to be an interesting approach in molecular modeling . This is a simple alignment-based approach based on the understanding that 2D chemical structure images contain relevant topochemical and topostructural information, useful for modeling chemical, physicochemical and biological properties of molecules (Antunes, Freitas, & Rittner, 2008;Bitencourt & Freitas, 2008;Cormanich, Freitas, & Rittner, 2011;Duarte, Barigye, da Mota, & Freitas, 2015;Duarte, Barigye, & Freitas, 2014;Freitas, 2006;Freitas, da Cunha, Ramalho, & Goodarzi, 2008;Goodarzi, Freitas, & Ramalho, 2009;Guimarães, Mota, Silva, & Freitas, 2014;Silla et al, 2011).Viewed from digital image processing standpoint, an image is basically comprised of an array of pixels arranged to yield different patterns according to a desired chemical image. Consequently, in order to employ these images as molecular descriptors, the numeric values for these pixels are considered.…”

Improvement of Multivariate Image Analysis Applied to Quantitative Structure–Activity Relationship (QSAR) Analysis by Using Wavelet‐Principal Component Analysis Ranking Variable Selection and Least‐Squares Support Vector Machine Regression: QSAR Study of Checkpoint Kinase WEE1 Inhibitors

“…Three‐dimensional QSAR studies have been previously performed to this class of compounds through CoMFA and CoMSIA (4), giving highly predictive models; thus, they serve as reference methods for comparison with our approach. Multivariate image analysis descriptors have shown to contain chemical information (23) and demonstrated high correlation with several bioactivities (6–15). Bilinear (traditional) PLS is often applied for such regression and, in a first approach, was used to calibrate the 73 × 63012 X ‐matrix with the activities column vector.…”

“…The bioactivities of a series of 4‐phenylpyrrolocarbazole derivatives, potentially useful as WEE1 inhibitors, have been modeled through molecular docking and 3D‐QSAR analyses, yielding high correlation between structural information and the inhibitory potency (4). Another QSAR strategy, which is based on the use of images (2D chemical structures) as explanatory variables, called multivariate image analysis applied to (MIA) QSAR, has provided several predictive QSAR models, without the need for conformational screening and 3D alignment (6–15). The correlation between such MIA descriptors and bioactivities is usually performed through partial least squares (PLS) regression (16), when a two‐way array is considered as regressors block, and multilinear PLS (N‐PLS) (17), applied to N ‐way arrays; however, alternative tools should be invoked when non‐linearity is present.…”

QSAR analysis of immune recognition for triazine herbicides based on immunoassay data for polyclonal and monoclonal antibodies

“…The quantitative structure-activity relationship (QSAR) is widely used to study the immune recognition of different classes of toxic food contaminants and veterinary drugs, including pesticides, etc. [11–14]. The work by Yuan and co-authors reported an immunoassay analysis of triazines; however, on only a small set of 11 compounds [9].…”