Fitting the ‘parity curve’: application of random regression to litter size in pigs

E.B. Sell-Kubiak, E.F. Knol, H.A. Mulder

Research output: Chapter in Book/Report/Conference proceedingAbstract


The objective of this study is to analyze the genetic background of variation in reproductive performance between parities of a sow with application of random regression. Performance of a sow (e.g. litter size) changes with the parity. Usually, the 1st and 2nd parity sows have smaller litters than sows in their 3rd to 5th parity. After parity 6, the performance declines again. The shape of litter size across parities closely resembles the lactation curve observed in dairy cattle. Fitting the ‘parity curve’ could help reducing differences between parities of a sow, preferably, keeping only the first parity with the lowest performance. Litter size data of Large White sows from commercial farms were provided by Topigs Norsvin. The records of 246,799 litters (total number born) came from 53,794 sows with at least two observations in parities 1-10. The analysis was performed in ASReml 4.1. Firstly, the 3rd order polynomials were selected as the best fit for the fixed parity curve. Secondly, the comparison of 3 models with respect to order of polynomials for random effects, indicated that the best fit was the model with 3rd order polynomials for both additive genetic and permanent environmental effects with simultaneous accounting for heterogeneity of residual variance per parity (10 levels). The heritability estimates were increasing with the lowest value in parity 1 with 0.11 and the highest value in parity 9 with 0.21. Genetically, parity 1 is the most different from parities 7-10 (rg from 0.56 to 0.60), whereas it is most similar to parities 2 (rg=0.96) and 3 (rg=0.85). Already the 2nd parity has correlations between 0.72 to 0.96 with other parities, whereas parities 3 to 10 are genetically almost identical traits with rg from 0.84 to 0.99. The non-unity genetic correlations between parities 1 and 2 with parities 3-10 show existence of genetic variation to genetically change the parity curve, e.g. to make litter size more equal among parities.
Original languageEnglish
Title of host publicationBook of Abstracts of the 68th Annual Meeting of the European Federation of Animal Science
Place of PublicationWageningen
PublisherWageningen Academic Publishers
ISBN (Electronic)9789086868599
ISBN (Print)9789086863129
Publication statusPublished - 2017
EventEAAP 68th Annual Meeting of the Animal Science (EAAP) - Tallinn, Estonia
Duration: 28 Aug 20171 Sept 2017

Publication series

NameBook of abstracts
ISSN (Print)1382-6077


OtherEAAP 68th Annual Meeting of the Animal Science (EAAP)


Dive into the research topics of 'Fitting the ‘parity curve’: application of random regression to litter size in pigs'. Together they form a unique fingerprint.

Cite this