Sang Van Nguyen defended his thesis on November 26th 2010

Ane Gro Siri Skjelfjord

Genetic studies on improvement of striped catfish (Pangasianodon hypophthalmus) for economically important traits

The aim of this study was to find non-invasive methods for measuring fillet weight, fillet yield and fillet fat on striped catfish (Pangasianodon hypophthalmus), to examine the magnitudes of genetic variance and covariance and potential genotype by environment interaction of economically important traits and finally direct selection response for body weight and correlated response for other traits. As part of the overall aim, modeling of fillet weight and fillet yield on body measurements and fillet fat with Distell Fish Fatmeter measurements were conducted. The final prediction equations achieved high correlations between predicted and observed fillet weight (0.93), fillet yield (0.86) and fillet fat (0.85), with the corresponding low biases of 1.4, 1.1 and 3.9%, respectively. Body weight, fillet weight, fillet yield, fillet fat and fillet colour were recorded in F2 of both populations 1, tested in the research station pond, and population 2, tested in three production systems; river-net fence, open-river pond and research station pond. Only body weight was recorded in F3 of population 1, and only in research station pond. Moderate to high heritability was obtained for body weight (0.21-0.52) and fillet weight (0.19-0.53), while low to medium heritability was found for fillet yield (0.02-0.09), predicted fillet fat (0.03-0.05) and fillet colour (0.04-0.20). Genetic correlation was positive and high between body weight and fillet weight (0.95-0.96), positive and moderate to rather high between these traits and predicted fillet fat (0.41 and 0.68-0.76, respectively), and low between these traits and fillet colour. The current proposed breeding goal traits are thus likely to be body weight/fillet weight and fillet fat, with fillet colour added later.

Genotype by environment interaction, measured as the genetic correlation of the same trait in different environments, was estimated. GxE interaction existed for all analysed traits in at least one pair of test environments (r = 0.57-0.83). With the average size of the genetic correlations for analysed traits being 0.69 between open-river pond and research station pond, the production being predominantly in the open-river ponds (80%) and the fact that the largest heritabilities found in this environment, it is concluded that testing and selection should be initially carried out in open-river ponds, or eventually that this environment is mimicked in research station pond. Alternatively the breeding program should test all full-sib families in the two largest environments, open-river pond and research station pond, for subsequent selection of the most stable genotypes across environments. The last alternative breeding strategies are to test genotypes in all relevant environments or to have one breeding program for each environment.

Selection response, estimated as the difference between least-squares mean of the selected group and the control group, for the trait increased body weight based on individual phenotypes. This was done over the first two generations in two populations. Substantial direct realised selection responses for body weight (4.6-12.4%) were found in both populations and they were significantly different from zero in two out of four instances. Realised heritabilities of 0.28-0.38 for body weight correspond well with the previously found heritability estimates. Correlated realised selection responses for fillet weight (4.5-12.0%) were also substantial and significantly different from zero, and with the same trend as for that with body weight, reflect the considerable heritability and high genetic correlation to body weight.

It is recommended that future works should include the application of optimum contribution selection to maximise the genetic gain, establishing genetic links among populations to uniform improved broodstock and large scale dissemination through multiplier network, testing of new economically important traits, such as salinity tolerance and disease resistance, and eventually also application of genomic selection.

Updated: 29.11.10
Printerfriendly version

Del med en venn:

Department of Animal and Aquacultural Sciences

P.O. Box 5003
N-1432 Ås

Phone: +47 64 96 51 00
Fax: +47 64 96 51 01

E-mail: iha@umb.no

Webmaster: Janne Karin Brodin