- Female reproduction traits play an important role in the economy of breeding companies with the sale of fertilized eggs.
- Results: A pedigreed population of 1343 trout were genotyped for 57,000 SNP markers and phenotyped for seven traits at 2 years of age: spawning date, female body weight before and after spawning, the spawn weight and the egg number of the spawn, the egg average weight and average diameter.
- The female body weight was not genetically correlated to any of the reproduction traits.
- Spawn weight showed strong and favourable genetic correlation with the number of eggs in the spawn and individual egg size traits, but the egg number was uncorrelated to the egg size traits.
- The genome-wide association studies showed that all traits were very polygenic since less than 10% of the genetic variance was explained by the cumulative effects of the QTLs: for any trait, only 2 to 4 QTLs were detected that explained in-between 1 and 3% of the genetic variance.
- The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material.
- Full list of author information is available at the end of the article.
- Breeding goals, i.e., the number, the nature and the importance given to the selected traits, are an essen- tial feature of a breeding program because this deter- mines the direction and extent of genetic trends in the population under selective breeding.
- However these traits play an important role in the economics of a breeding company because of the sale of eyed eggs.
- A high reproductive capacity is also one of the main levers for the genetic improvement of the species.
- The aims of the study were to estimate genetic parameters, to identify quantitative trait loci (QTL) and to assess the ef- ficiency of genomic selection (GS) compared to pedigree-based BLUP selection for female reproduction traits in one of these French rainbow trout selected line..
- Concerning this last objective, change in selection effi- ciency was investigated according to two main factors of variation: the size of the reference population and the degree of kinship between reference and candidate pop- ulations for selection.
- Our estimates of correlations for the two measures of female body weight with (FW) or without (PW) consid- eration of the spawning and coelomic liquid weights showed that the two measures were essentially describ- ing the female own weight since we estimated a pheno- typic correlation between FW and PW of 0.96 with a genetic correlation close to unity.
- On the contrary, ED and EW could be consid- ered as two different measures of a unique biological egg size trait, those two traits being associated in a very simi- lar manner to all the other traits in the analysis.
- The female body weights were also not genetically correlated to any of the egg size traits (EW and ED) or egg quantity traits (EN and SW).
- Considering FW as a covariate in the.
- With the only exception of SD, no QTL explained over 3% of the genetic variance (Table 2).
- of genetic variance explained by all the SNPs included in the QTL credibility interval.
- Only 2 to 4 QTLs were de- tected that explained at least 1% of the genetic variance for any of the five reproduction traits (Table 2).
- Ten of the twelve QTLs that were at 1% chromosome-wide sig- nificant under GBLUP analysis corresponded to QTLs with strong evidence under the Bayesian approach.
- In total, the Bayesian ap- proach allowed the detection of 17 QTLs that explained at least 1% of the genetic variance, but only nine had a very strong evidence (logBF >.
- The GBLUP accuracy was higher than the BLUP accuracy in most of the 40 validation samples (Fig.
- of the samples failed to respect this rule in average over all the traits.
- In 60% of the 40 samples, GBLUP was more accurate than BLUP for any of the six traits.
- loss of accuracy of GBLUP was less than 10% in 80% of the remaining samples..
- The estimated heritability values for FW and PW in the present study are within the range of values reported in recent studies in the same species for female body weight between 13 and 25 months [17–19].
- the full-sib family effect was very significant ac- counting for 22, 13 and 10% of the phenotypic variance for the three lines, respectively.
- However our results do not really validate this assumption since FW and PW do not appear to be genetically correlated - neither positively nor negatively - to any of the reproduction traits (SD, SW, EN, EW or ED).
- Nevertheless, based on the functional information given in the human gene database GeneCards® [21, 22] and described pheno- types in mutant mice or worms, we were able to propose five candidate genes for female reproduction traits (3 for SD, 1 for ED and 1 shared by EN and SW).
- No pheno- types were described for these candidate genes in the zebrafish database ZFIN [23].
- Fitness traits such as spawning date and body weight are major factors in the life history of salmonid fishes..
- Despite the fact that some recent QTLs studies have fo- cused on growth traits in rainbow trout [17, 19] the only significant SNP we detected for female body weight was not in the vicinity of any QTL regions reported for trout body weight.
- of the genetic variance under BayesC approach.
- Nevertheless, as far as we know, no QTL for SD in salmonid species has been reported in the neighbor- hood of the highly significant SNP we observed on Omy6.
- Each of them explained between 1.2 and 1.9% of the genetic variance of SD.
- Interestingly, we can hypothesize that two of the four QTLs detected for SD may be associated to abnormal eye morphology and defects in vision that may render trout less sensitive to the photoperiod stimuli..
- The QTL on Omy2 explained 3.2 and 3.0% of the genetic variance for EN and SW, respectively.
- The QTL on Omy12 ex- plained about 2.5–2.7% of the genetic variance for each trait.
- For EN, a third QTL was detected on Omy8 that explained 1.4% of the genetic variance.
- For SW, a third QTL was detected on Omy2 that explained 1.9% of the genetic variance and a last QTL was detected on Omy1 that explained 1.3% of the genetic variance.
- This PHC1 gene is a homolog of the Drosophila polyhomeotic gene, which is a member of the Polycomb group of genes.
- No obvious candidate gene could be proposed in this large QTL region that explained 2% of the genetic variance for EW..
- Regarding EW, there was a strong evidence for a dis- tinct QTL explaining 1.2% of the genetic variance, in the region spanning from 60.498 Mb and 64.633 Mb on Omy1 with the peak SNP very close to 64.633 Mb in an uncharacterized protein (LOC110527930).
- No candidate gene could be proposed within this QTL region, but in the close vicinity of this QTL region, let us mention the presence of the prkg2 gene (located between 64.660 and 64.676 Mb on Omy1) whose role is important in oocyte maturation in mammals and zebrafish [32].
- Regarding ED, there was evidence for another QTL on Omy1 that explained nearly 2% of the genetic variance in the region spanning between 70.848 Mb and 71.813 Mb.
- The peak SNP (at 71.813 Mb) is close to the pos- ition Mb) of the WAPLA gene (wings apart-like protein homolog) which is a straightforward gene candidate for explaining this last QTL on Omy1..
- In the literature, several studies in salmonids reported moderate to strong gains in accur- acy.
- 11% to 110%) for genomic selection compared to BLUP selection of depending on the genetic architecture of the traits and the size of the reference populations.
- In- creasing by 60% the size of the training population (starting from 670 individuals) in a fish line whose ef- fective population size is estimated around 50 [6], in- creased the accuracy of GEBVs by 6% to 11% depending on the trait considered.
- It is well known that close relation- ships between animals in the training and validation sets increase the accuracy of genomic predictions compared to the ones derived for an independent validation popu- lation [41].
- Under scenarios T1 and T2, our estimates of GS accuracy are close to the theoretical estimates de- rived from Goddard [39]’ formula, which makes sense since one of the basic assumption beyond his formula is that GS accuracy comes only from linkage disequilib- rium across the whole population and not from any link- age association and family structure.
- This may be a strong issue to correctly predict genetic trends or performing optimal multitrait index selection since the magnitude of the in- flation varied a lot across traits.
- Nevertheless, when per- forming selection within cohort based on a training population including sibs of the candidates, it does not appear to be a problem..
- In our study, all female reproduction and weight traits were moderately heritable with spawn weight showing strong and favorable genetic correlations with number of eggs in the spawn and individual egg size traits (egg average diameter or weight).
- On the contrary, number of eggs in the spawn was uncorrelated to egg size traits and female body weight (just before or after spawning) was not genetically correlated to any of the reproduction traits.
- Only six QTLs over the 19 identified across the five traits studied explained at least 2% of the genetic variance.
- Phenotypes were collected at 2 years of age in females from two successive cohorts produced in 2014 and 2015, hereafter named C1 and C2, and composing the 9th gen- eration of selection of the trout breeding company.
- Those cohorts constituted the broodstock of the company and came from two related paternal cohorts S1 and S2 and from different groups of dams D1 and D2 of the same maternal cohort produced in 2011 (Fig.
- The animals used in the study were reared at the French farm “Viviers de Sarrance” (Pisci- culture Labedan, 64,490 Sarrance, France).
- Raw phenotypes collected were the weight of the ready-to-spawn female (FW), its post-spawning weight (PW), the weight of the total egg mass hereafter called the spawn weight (SW), the length of 50 eggs aligned along a graduated rule, the weight and the number of eggs in a sampling spoon of 2.5 ml, the spawning week number in the calendar year and the presence of over- mature eggs in the spawn.
- EW was derived as the ratio of the weight of eggs to the number of eggs in the sample of 2.5 ml..
- SD corre- sponded to the rank of the week number within the.
- In total, the phenotypes of 1517 fish were considered in the study (Table 4)..
- Most of the fish have their parents also geno- typed for the 57,501 SNPs.
- For any trait i among the six reproduction traits con- sidered in the study, the following statistical model was considered to describe the vector of performance y i of the 1346 fish:.
- Table 4 Summary statistics of 2-year female reproduction and weight traits in the rainbow trout broodstock.
- For the traits FW, SW, EN and ED, an additional fixed effect due to the presence of overmature eggs was significant and there- fore considered in the models.
- Tracing back over 8 generations the pedigree of the 1346 phenotyped fish, the vector u i corresponded to the breeding values of 15,265 individuals related through the pedigree relationship matrix A..
- (1), but only the performance of the genotyped animals can be integrated into a conventional GBLUP analysis:.
- A region of the genome was considered to be a QTL when the -log 10 (p-value) for a SNP of this region was equal or greater than 5.0 (which corresponds to a chromosome-wide significance threshold of 1% derived as -log 10 (0.01/(n/30)) after Bonferroni correction with n = 29,799 the total number of SNPs included in the analysis).
- Therefore, a general linear mixture model was defined in which a fraction π of the 30 K SNPs was assumed to have a non-zero effect at each cycle of the MCMC algorithm.
- Convergence was assessed by visual inspection of plots of the posterior density of gen- etic and residual variances and by deriving high correla- tions (r >.
- 0.99) between GEBVs estimated from different chains of the MCMC algorithm.
- By trial-and-error, this π value was considered as a good compromise in our vari- able selection algorithm between the high degree of pol- ygeny of the quantitative traits under study and the limited number of individuals (n ~ 1300) in our dataset that led to consider p = 300 <.
- involves π and P i , the probability of the ith SNP to have a non-zero effect: BF ¼ P π=ð1 i =ð1.
- As proposed by Kass and Raftery [53], the logBF was computed as twice the natural logarithm of the BF and the threshold logBF ≥ 6 was used for defining evidence for a QTL.
- [54], a cred- ibility interval was built around the peak SNP integrating to the QTL region the SNPs with logBF ≥ 3 that were lo- cated close to the peak SNP using a sliding window of 1 Mb on both sides of the peak SNP.
- 8 corresponded only to a putative QTL unless the QTL region explained at least 1% of the genetic variance;.
- The inflation coefficient was derived as the regression coefficient of the corrected phenotypes on the (G)EBVs.
- In the absence of selection bias, this coeffi- cient is expected to be equal to 1.
- For the T+ and T- scenarios, accuracy of selec- tion and inflation coefficient were derived as the mean over the 40 replicates of the correlation and regression coefficients previously described between (G) EBV and corrected phenotypes for the validation population..
- We thank the staff of the French breeding company Viviers de Sarrance for their participation in the fish rearing and phenotyping and also for providing biological samples..
- JDA performed the curation, the analysis and interpretation of the data and was involved in the script writing, the interpretation of the results and the writing of the original draft.
- RM and SBF were involved in the sampling of the animals, the data investigation and the supervision of the work.
- AB was involved in the curation and the investigation of the data.
- AAP performed the sampling and phenotyping of animals and was involved in the data investigation.
- CP managed the animal genotyping and was involved in the data investigation.
- DG was involved in the funding acquisition and conception of the work.
- PH managed the funding acquisition and the administration of the project and was involved in the conception and supervision of the work.
- MD was involved in the conception and supervision of the work.
- FP was involved in the conception, choice of design and statistical methodology, results interpretation and supervision of the work, and wrote the original draft.
- D ’ Ambrosio ’ s phD thesis, the animal phenotyping and genotyping, the data analysis and the communication of the results.
- The funding bodies played no role in the design of the study and collection, analysis, and interpretation of data and in writing the manuscript..
- The data can be made available for reproduction of the results from Florence Phocas (
[email protected]) and Ana Acin-Perez (
[email protected]) on request via a material transfer agreement and with permission of the breeding company « Viviers de Sar- rance.
- The accession numbers for the rainbow trout reference sequence assembly and the corresponding gene bank assembly were GCF and GCA in the National Center for Biotechnology Information database that can be accessed at the weblink: https://www.ncbi.nlm.nih.gov/.
- Art- icle 1.5 in the EU Directive 2010/63/ EU on the protection of animals used for scientific purposes excludes practices undertaken for the purposes of recognised animal husbandry and identification from the scope covered by the Directive 2010/63/ EU.
- A comprehensive survey on selective breeding programs and seed market in the European aquaculture fish industry.
- Aquaculture genomics, genetics and breeding in the United States:.
- A genetics analysis of the performance of three rainbow trout broodstocks.
- Extension of the bayesian alphabet for genomic selection