Initial Study Of Revised Expressiongs On Early Growth Performance Of Lean Meat Type Breeding Swine

This text used the typical lean meat pig Duroc, Landrace and Large white as the research objects, carried on the performance testing at early days and whole period(30-100 kg), the early testing data is collected once for every two days, which was overall continuous data. The collected data came from five farms of breeding swine in Hubei province, which contained 678 pigs data and had more than 20,000 testing information about earlier period. This text applies variety no linear model, such as the Logistic, Gompertz, Saturation, Quadratic and Polynomial etc., to take the various ways regress analysis for growth rules of early age and weight, weight and back fat. The author had draw up the revised expressions of 50 kg weight age and back fat for the three breeds, which has its biology meaning and representation, and had investigated the most suitable regress equation to describe the early growth traits.Took SAS software as the main tool to calculate and analyse the testing data. The results indicated: Logistic, Gompertz and the Saturation is more suitable for regress analysis of the relation about age and weight, isn't suitable for weight and back fat. In the three breeds with different sex, all correlation indexes values excessed 0.9 average, and attained the extreme difference(P＜0.01), the Gompertz equation has biology meaning more than the Logistic equation, the results expressed that growth rule in three breeds is different. The growth curve inflexion of Landrace is the most smallest, and large white is the biggest in the three breeds. It elucidates that the large white has good potential growth and It's maturity stage is later compared with the Landrace. The Saturation equation and line equation are more suitable to research the back fat with weight, all correlation indexes values and decision coefficient are very high, and all regress equations attained the extreme difference(P＜0.01), on the other hand the fitting regress curve of all breeds looks like a straight line on weight and back fat. The equation has biology meaning to predict the thickness of back fat in a premise where the weigh of pig has a extreme. Furthermore, straight line has the more advantage to establish revised expressions compared with the Saturation equation. The text took the lead in applying Saturation growth model to carry on the research of the pig growth regulation, and the regress effect is good, the goodness was great. Because of the t variable has no Polynomial or no linear switch, so that the saturation equation has the advantage to establish revised expressions. The result indicated that the revised expressions for age in 50 kg standard weight has explicit biology meaning and accuracy by establish with the Saturation. This text would like to take continuous testing data as foundation, and established the commonly revised expressions for early growth traits, which are suitable for different sex and breeds. These expressions have function to correct and predict for continuous testing data of pigs at 40-60 kg weight, and have the same function with those overseas. This test has established two kinds of expressions of back fat, which have its biology meaning, the accuracy is completely same by experience certificate. Get across the simple correlation analysis(CORR) and canonical correlation analysis(CANCORR), using the early adjusted value and whole period adjusted value, the all results reached the extreme difference (P＜0.01). It proved the expressions are suitable, and the early testing technology is accuracy and feasible. Additional, the Spearman's rank correlation reached the extreme difference (P＜0.01) for total selection index. It shows, although the order are different, that they have very high consistency at certain choice strength and have the field and latent applied value to instruct selection in farm.The initial research of this thesis matches the demand of the development of breeding swine industry. The author would like do some foundational work for early growth traits testing and genetic evaluation of lean meat breeding swine. The author hope that the early growth traits and genetic evaluation can be used at certain scope, and make the contribution to investigate our country early genetic evaluation of breeding swine by combining with nucleus herd testing currently in the future.