Abstract: In order to explore the relationship between the structure and bioactivity of zinc-chelating peptides, quantitative structure-activity relationship (QSAR) models were established. A new method called two-terminal position numbering was proposed to describe the structures of 56 synthetic zinc-chelating peptides with different lengths. Then, these peptides were statistically analyzed using 18 amino acid descriptors and partial least squares regression. Results showed that the correlation coefficients of the QSAR models based on 5 amino acids descriptors FASGAI, Z, HESH, C and ST met the requirements. FASGAI was found to be the best among these descriptors (R2 = 0.827 3, Q2 = 0.602 2, root mean square error of estimation = 0.168 6, Q2 ext = 0.717 2, and root mean square error of prediction = 0.255 8). Further analysis of the FASGAIbased model revealed that the influence of amino acid positions in the peptides on their zinc-chelating activity was C3 > N3 > C1 > N1 > N2 > C2. Meanwhile, the bulky properties of amino acid residues influenced the zinc-chelating activity of the peptides. These models have the potential to provide new ideas to explore the quantitative structure-activity relationships of zinc-chelating peptides.

Key words: rapeseed source, zinc chelating peptide, quantitative structure-activity relationship