作者: Brij Kishore Sharma
DOI: 10.4172/2161-0444.1000134
关键词: Applicability domain 、 Data mining 、 Matrix (chemical analysis) 、 Linear regression 、 van der Waals force 、 Chelation 、 Biological system 、 Eigenvalues and eigenvectors 、 Weighted distance 、 Quantitative structure–activity relationship 、 Chemistry
摘要: The MMP-13 inhibition activity of non-zinc-chelating compounds has been quantitatively analyzed in terms chemometric descriptors. statistically validated quantitative structure-activity relationship (QSAR) models provided rationales to explain the these compounds. descriptors, identified through combinatorial protocol multiple linear regression (CP-MLR) analysis, have highlighted role 3-path Kier alpha-modified shape index (S3K), complementary information content 1-order neighborhood symmetry (CIC1), eigenvalue sum from mass weighted distance matrix (SEigm), lowest n. 6 Burden / by atomic van der Waals volumes (BELv6) and polarizabilities (BELp6), 3-order topological charge (GGI3) functionality, R--CR--R (C-025). From models, it appeared that descriptors S3K, BELv6, BELp6 SEigm make positive contribution their higher values are conducive improving MMP- 13 a compound. On other hand, CIC1, GGI3 C-025 render detrimental effects activity. Therefore, absence lower CIC1 would be advantageous. PLS analysis further corroborated dominance CP-MLR Applicability domain revealed suggested acceptable predictability. All within applicability proposed were evaluated correctly.