Optimizing Genomic Prediction of Host Resistance to Koi Herpesvirus Disease in Carp

Christos Palaiokostas, Tomas Vesely, Martin Kocour, Martin Prchal, Dagmar Pokorova, Veronika Piackova, Lubomir Pojezdal, Ross Houston

Research output: Contribution to journalArticlepeer-review


Genomic selection (GS) is increasingly applied in breeding programs of major
aquaculture species, enabling improved prediction accuracy and genetic gain compared to pedigree-based approaches. Koi Herpesvirus disease (KHVD) is notifiable by the World Organization for Animal Health and the European Union, causing major economic losses to carp production. GS has potential to breed carp with improved resistance to KHVD, thereby contributing to disease control. In the current study, Restriction-site Associated DNA sequencing (RAD-seq) was applied on a population of 1,425 common carp juveniles which had been challenged with Koi herpes virus, followed by sampling of survivors and mortalities. GS was tested on a wide range of scenarios by varying both
SNP densities and the genetic relationships between training and validation sets. The accuracy of correctly identifying KHVD resistant animals using GS was between 8 and 18% higher than pedigree best linear unbiased predictor (pBLUP) depending on the tested scenario. Furthermore, minor decreases in prediction accuracy were observed with decreased SNP density. However, the genetic relationship between the training and validation sets was a key factor in the efficacy of genomic prediction of KHVD resistance in carp, with substantially lower prediction accuracy when the relationships between the training and validation sets did not contain close relatives.
Original languageEnglish
Article number543
JournalFrontiers in genetics
Publication statusPublished - 12 Jun 2019


  • KHVD
  • carp
  • RAD-seq
  • genomic selection
  • aquaculture breeding

Fingerprint Dive into the research topics of 'Optimizing Genomic Prediction of Host Resistance to Koi Herpesvirus Disease in Carp'. Together they form a unique fingerprint.

Cite this