Comparison of Different Non-Linear Models for Describing Plasma Lysozyme Activity in Quail

Abstract

Lysozyme activity is one of the nonspecific immunity parameters measured by changing the amount of adsorption at different times. The objective of the present study was to compare five non-linear models including Gompertz, Richards, Logistic, Lopez, and Weilbull to describe the cumulative plasma lysozyme activity in quails. In total 1364 plasma samples (1004 females and 360 males) were collected and the cumulative lysozyme activity was calculated by turbidimetric method assay in Micrococcus luteus. The goodness-of-fit of models was compared according to different criteria of Maximum log-likelihood, Akaike information criterion, Mean square error, and Bayesian information criterion. The results showed that the Gompertz model was the best model for describing of decreasing cumulative pattern of lysozyme activity in female and male quails and provided satisfactory predictions of lysozyme activity at different times (30, 60, 90, 120, 150, 180, 210, 240, 270, 300 seconds). The parameters of all models were higher in females than males except for the k parameter which was greater in the males. Male quails had higher values for time and lysozyme activity than females at inflection points, whereas the absolute growth rate in 30, 150, and 300 seconds was predicted higher in female quails. In conclusion, the Gompertz model can be used accurately to evaluate cumulative lysozyme activity patterns in both sexes of quails.

Keywords

• quail
• immunity
• non-linear models
• inflection point