Comparison of single- and multi-trait approaches to identify best wild candidates for aquaculture shows that the simple way fails.

In agriculture, diversifying production implies picking up, in the wild biodiversity, species or populations that can be domesticated and fruitfully produced. Two alternative approaches are available to highlight wild candidate(s) with high suitability for aquaculture: the single-trait (i.e. considering a single phenotypic trait and, thus, a single biological function) and multi-trait (i.e. considering multiple phenotypic traits involved in several biological functions) approaches. Although the former is the traditional and the simplest method, the latter could be theoretically more efficient. However, an explicit comparison of advantages and pitfalls between these approaches is lacking to date in aquaculture. Here, we compared the two approaches to identify best candidate(s) between four wild allopatric populations of Perca fluviatilis in standardised aquaculture conditions. Our results showed that the single-trait approach can (1) miss key divergences between populations and (2) highlight different best candidate(s) depending on the trait considered. In contrast, the multi-trait approach allowed identifying the population with the highest domestication potential thanks to several congruent lines of evidence. Nevertheless, such an integrative assessment is achieved with a far more time-consuming and expensive study. Therefore, improvements and rationalisations will be needed to make the multi-trait approach a promising way in the aquaculture development.

Aging three-spined sticklebacks Gasterosteus aculeatus: comparison of estimates from three structures.

In order to assess the accuracy and reliability of age estimates from calcified structures in the three-spined stickleback Gasterosteus aculeatus, we evaluated intra and inter-reader repeatability from three structures: otoliths, gill covers and pelvic spines). Average age estimates were also compared between the structures. The overall intra-reader repeatability of age estimates were highest for otoliths (69%), lowest for gill covers (53%) and intermediate for spine cross-sections (63%). Although four of the seven readers had the highest intra-reader repeatability score for spine cross-sections, the inter-reader variance in this structure was much higher than in others. Otoliths were the easiest in terms of their pre-analysis treatment and exchange of materials (as digital images) between readers. In addition, otoliths are more well-studied compared with the other structures with respect to their development through ontogenesis; hence, age estimates based on otoliths should be the most reliable. Therefore, our recommendation is that whenever possible, analysis of otoliths should be the preferred approach for aging G. aculeatus.