Predicting the Activity of ACE Inhibitory Peptides with a Novel Mode of Pseudo Amino Acid Composition

Mao      Shu; Xiaoming      Cheng; Yunru      Zhang; Yuanqiang      Wang; Yong      Lin; Li      Wang; Zhihua      Lin

Abstract

In this study, physicochemical scale (P-scale), was recruited as a novel set of physicochemical descriptors derived from component analysis on four short of physicochemical properties variables (hydrophobic, electronic, steric and hydrogen bond contribution) of 20 coded amino acids, By using partial least squares (PLS), we applied P-scale for the study of quantitative structure – activity relationship models (QASRs) on three angiotensin I converting enzyme (ACE) inhibitory peptides datasets (58 dipeptides, 55 tripeptides, and 50 tetrapeptides).The results of QSARs were superior to that of the earlier studies, with correlation coefficient (r2) and cross-validated(q2) equal to 0.902, 0.86; 0.985, 0.951 and 0.872, 0.77, respectively. By analysis, hydrophobic and steric properties of ACE-inhibitory peptide sequences play important roles in their bioactivities, and novel peptide sequence could be designed based on these properties of the amino acid residues. These results showed that P-scale descriptors can well represent the peptide sequence. Furthermore, the robust models show that P-scale descriptors can be further expanded for polypeptides and can serve as a useful quantitative tool for the rational drug design and discovery.

Keywords: angiotensin I-converting enzyme-inhibitory peptides, genetic algorithm, partial least square, physicochemical scale(P-scale), quantitative structure, –, activity relationship, ACE leads, bradykinin, QSAR, oxytocin vasopressin analogues, PCA, Peptide structural characterization, Hydrophobic, PseAAC, Variables Selection, GAangiotensin I-converting enzyme-inhibitory peptides, genetic algorithm, partial least square, physicochemical scale(P-scale), quantitative structure, –, activity relationship, ACE leads, bradykinin, QSAR, oxytocin vasopressin analogues, PCA, Peptide structural characterization, Hydrophobic, PseAAC, Variables Selection, GA