Zumbach, Birgit: Schätzung von Kreuzungsparametern unter besonderer Berücksichtigung von epistatischen Effekten und einer Optimierung des Kreuzungszuchtverfahren beim Meerschweinchen in Bolivien


8. Summary

Estimation of Crossbreeding Parameters with Special Consideration of Epistatic Effects and Optimisation of Crossbreeding Procedures For Guinea Pigs in Bolivia

The present study investigates the guinea pig as a native animal for meat production in the Andean countries of Latin America. In the framework of rural development the guinea pig can contribute to a better meat supply and a source of income especially for small holders by improving breeding and management. The objective of this study is to estimate crossbreeding parameters, focusing on epistatic effects, from crossings between a small framed synthetic native line consisting of 29 Bolivian strains and an imported line from Peru selected for body weight, kept under semi-intensive/semi-extensive ”modern“ management conditions. Conclusions for the optimisation of breeding systems should also be derivated.

In the introduction the special biological characteristics of the guinea pig are presented briefly. The cultural and social meaning of the guinea pig, traditionally being more than only a meat resource, and the traditional management system are described. Possible difficulties when using ”modern“ breeding animals in traditional management systems are pointed out, and a global Knowledge Network in the area of investigation including traditional knowledge as well as the experiences of scientific groups on guinea pigs all around the world, is suggested.

The literature review consists of three chapters: In the first one gene effects in crossbreeding, especially the different kinds of epistatic effects in different genetic models and the term of heterosis are treated. Complementarity and gene effects in different generation levels are presented. The second chapter deals with planning and optimising of crossbreeding systems. The third one is a literature review of guinea pig reproduction and production performance in pure- and crossbreeding.

In the present study, data from 2857 first and second litters and 7745 progenies, stemming from the matings of 391 bucks with 1688 dams were analysed. The animals belong to a native synthetic Bolivian line, developed at the Universidad Mayor de San Simón, Cochabamba, Bolivia, and an imported Peruvian line as well as to 12 crossbreeding groups derived from these stocks: reciprocal F1s, reciprocal backcrosses, F2, pre-rotational groups and a synthetic line with 50% gene proportion of each purebred line.

The basic model for the estimation of crossbreeding parameters is the Jakubec model with three different kinds of epistatic effects (additive x additive interactions, additive x dominance interactions, dominance x dominance interactions) and the Dickerson model with the recombination loss as epistatic effect on the individual (progeny), maternal (individual), paternal (male mating partner) and grandmaternal (maternal) level. The theoretical estimability of each parameter was derived by describing the breeding groups as linear combinations of crossbreeding effects for the Jakubec and the Dickerson model. In the Jakubec model twelve out of the 20 described parameters are estimable and in the Dickerson model eight out of twelve.

In an example it is shown that the estimation of all theoretically estimable parameters by an unbiased regression procedure doesn’t make sense because of the existing inter-correlation between parameters (co-linearity). The number of parameters per genetic model was reduced to five, each model including one epistatic effect: Dickerson model gi, hi, ri, gm, hm and gnk, hnk, rnk, gi, hi; Jakubec model aa: gi, di, aai, gm, dm and gpr, dpr, aapr, gi, di; Jakubec model dd gi,


di, ddi, gm, dm and gpr, dpr, ddpr, gi, di (i: individual, m: maternal, pr: progeny). The synthetic native Bolivian line was considered as reference.

Breeding group means and crossbreeding parameters for single traits were analysed separately for each litter number in order to properly taking into account the differently influencing environmental effects. Weight and growth performance were evaluated using an animal model. Variance components were also estimated: Heritability estimates for weight traits from birth to the age of 84 days vary between 0.27 and 0.29. The estimates for the weight gain from birth to weaning (14 days after parturition) and from weaning to the age of 56 days are 0.19 and 0.16, respectively, while there doesn’t seem to be a genetic base for the weight gain from the 56th to the 84th day of age (h²=0.03).

The GLS breeding group means for the birth weight and the weight development to weaning show a clear maternal influence for the progeny of purebred dams, while pre-natal and post-natal weight development of crossbred dams’ progeny depend primarily on exterior environmental conditions which were quite different within and between breeding groups. Maternal and individual line differences are the prevailing crossbreeding effects until weaning: For the birth weight they measure about 30-35 g or 37% of the GLS estimate for the Bolivian line, with individual effects dominating in the first litter and maternal effects in the second. Negative epistatic effects (-8 to -17%) outline in both litters. In the first litter also negative maternal dominance effects (-4 to -6%) are significant. For the weaning weight only line differences with a magnitude of about 30% can be detected. Survival rate of the progeny from birth to weaning is about 91.5% in both litters, whereby year season effects play an important role.

In the weight development after weaning a clear differentiation of the breeding groups by Peruvian gene proportion can be recognised. The individual line differences are the dominating crossbreeding effects reaching a magnitude of 170 to 205 g or 50% (2nd litter) to 80% (1st litter) for the weight gain from weaning to the age of 84 days. In the 2nd litter also positive individual dominance effects (12-13%) can be detected. Negative maternal line differences (5%) outline in the first litter. Body weight at 56 and 84 days of age is characterised by the crossbreeding effects on the weight development until weaning and the weight gain after weaning. Nevertheless individual line differences are dominating in the body weight at the age of 56 days (36-37%) as well as in the age of 84 days (1st litter: 59%; 2nd litter: 37%). The maternal ones stemming from the weaning period can still be detected at the age of 84 days (7.5% in the 2nd litter). As non-additive effects only positive individual dominance effects differ significantly from zero.

A slaughter trial was conducted consisting of a sample of male animals at the age of 85 days. Carcass weight (280.5 g to 385.5 g) and backfat percentage (0.52% to 0.70%) increase with increasing Peruvian gene proportion, while the ratio between carcass length and carcass weight is declining. Only line differences in the magnitude of 37% for carcass weight and backfat percentage and of -19% for carcass length ratio, respectively, can be detected.

Parturition rates for the first and second litter average 96 and 90%, respectively. For the first parturition there is no difference between breeding groups, while the proportion of Peruvian dams having a second litter (80%) is substantially lower than that of others. Year season effects are of great importance in both litters. The dams’ survival rate until first parturition and after first parturition to the second averages 99% and 96%, respectively. Mortality of Peruvian dams after first parturition (10%) was markedly higher than that of others.

The interval between first mating and first parturition and between first and second litter average 75.4 and 71.2 days, respectively. Comparing the LS-breeding group means no genetic difference between breeding groups can be detected. Considering the number of litters per dam and half


year (: 2.46) there is a clear tendency for more progenies in mating types with Bolivian dams than in those with Peruvian dams. Crossbreeding parameter estimation shows clearly the existence of genetic effects. For the interval until first parturition positive line differences (5.2%) and negative dominance effects (-3 to -4%), both on the progeny level, can be detected. For the interval between first and second litter only line differences, on the progeny level (3.5%) and on the individual level (3.0%) are shown. The parameters for number of litters per half year correspond to those of the interval between first and second litter: They consist in negative line differences with a magnitude of -2.8% on both individual and progeny level, respectively.

Comparing the breeding group means for litter size at birth and at weaning, a genetic difference between breeding groups is not obvious. Crossbreeding parameter estimation for litter size at birth shows negative dominance effects on the progeny level in the first litter - in the second they outline - as well as negative line differences on the progeny level in the second litter. At weaning only the negative line differences in the second litter can be detected.

Litter weight, consisting of the components litter size and kid body weight is primarily determined by kid body weight. Mating types with Peruvian dams - with the exception of BBxPP - achieve highest litter weights, mating types with Bolivian dams lowest. Crossbreeding parameter estimation for litter weight at birth shows positive line differences (on the progeny and/or individual level) in the magnitude of app. 30% in both litters, whereby the negative additive progeny effects from litter size are compensated by the positive individual line differences in kid body weight. In the first litter negative dominance effects on the progeny level originating from the litter size component can be detected, too. Crossbreeding parameters for litter weight at weaning correspond mainly to those for litter weight at birth.

Productivity criteria such as number or weight of progeny per dam and half year comprise beside the primary reproduction traits as litter size or litter weight also parturition rate, litter frequency and survival rate of breeding animals and progeny. Because of the marked difference in dam body weight ranging from 640 g (Bolivian line) to 1040 g (Peruvian line), also metabolic dam weight which takes into account the different requirements of maintenance was considered as reference.

Bolivian dams excel in the number of progeny per dam and in a much higher degree per metabolic dam weight that of Peruvian ones from birth until 84 days after parturition during the first half year of production. The performance of crossbred dams, with the dam as reference, lies about in between, with reference to the metabolic dam weight, mating types with lighter dams are markedly superior to heavier ones. The mating type BBxPP shows exceptionally low performance.

Crossbreeding parameter estimation shows negative line differences (-11.6%) and dominance effects (-10 to -20%) on the progeny level for number of progeny per dam at birth. At weaning and 84 days after parturition additionally to the negative line differences on the progeny level negative individual ones (-8 to -10%) can be detected while non-additive effects are not significant. Taking the metabolic dam weight as reference, negative individual line differences (-33 to -36%) are dominating from birth until 84 days after parturition. On the progeny level they have the magnitude of -7 to -8%. Non-additive effects only outline as negative dominance effects on the progeny level for the number of total progeny born.

The progeny weight per dam at parturition in the first half year of production is primarily determined by the progeny’s birth weight, i.e. mating types with Bolivian dams are substantially inferior to those with Peruvian dams, except the mating type BBxPP for its very low performance in progeny number. At weaning, however, no genetic differences between breeding groups can be


recognised. 84 days after parturition mating types with a higher Peruvian gene proportion in total also achieve higher progeny weights. With reference to the metabolic dam weight, mating types with lighter dams (low Peruvian gene proportion) are superior to the others. After weaning highest progeny weights are achieved by mating types with a high Peruvian gene proportion in the sires and a low one in the dam.

Crossbreeding parameter estimation results in positive individual line differences (12%) for the progeny weight per dam and half year at birth. With reference to the metabolic dam weight those line differences are negative with a magnitude of -23%. At weaning no crossbreeding effect can be detected for the progeny weight per dam while per metabolic dam weight the negative individual line differences have risen to -30%. 84 days after parturition positive line differences on the progeny level (23%) can be detected for both references (dam and metabolic dam weight). While with reference to the dam no other effects are significant, the negative additive individual effects have reached a magnitude of -40% with relation to the metabolic dam weight.

Finally the genetic models and the results are discussed in a global context and recommendations are derived for breeding systems and selection for the investigated lines under the specific seasonal and economical conditions prevailing at Cochabamba, Bolivia.

[Titelseite] [Danksagung] [Abkürzungsverzeichnis] [1] [2] [3] [4] [5] [6] [7] [8] [Bibliographie] [Selbständigkeitserklärung] [Anhang]

© Die inhaltliche Zusammenstellung und Aufmachung dieser Publikation sowie die elektronische Verarbeitung sind urheberrechtlich geschützt. Jede Verwertung, die nicht ausdrücklich vom Urheberrechtsgesetz zugelassen ist, bedarf der vorherigen Zustimmung. Das gilt insbesondere für die Vervielfältigung, die Bearbeitung und Einspeicherung und Verarbeitung in elektronische Systeme.

DiDi DTD Version 1.1
a subset from ETD-ML Version 1.1
Zertifizierter Dokumentenserver
der Humboldt-Universität zu Berlin
HTML - Version erstellt am:
Tue Mar 16 17:16:09 1999