At present, aquaculture is the fastest growing sector of animal food production and holds great potential as a sustainable solution for world food security. The ability to control sex is one of the most important factors for the commercialisation and efficient propagation of fish species, due to influences on reproduction, growth and product quality. Accordingly, there is a large body of research that targets sexual development in commercially important species in an attempt to understand and control fish sex and reproductive function. In this review, we provide an introduction to sex determination and differentiation in fish, including the genetic, epigenetic and environmental factors that can influence fish sex ratios. We also summarise the major approaches used to control sex in fish and discuss their application in commercially important species. Specifically, we discuss the use of exogenous steroid hormones, chromosome ploidy, environmental manipulations, sex-linked genetic markers, selection for altered sex ratios, and transgenics and comment on the challenges associated with controlling sex in a commercial environment.

Epigenetics is involved in sex differentiation of gonochoristic and hermaphroditic fish species, whereby two genes dmrt1 (pro-male) and cyp19a1 (pro-female) are known to play major roles. Barramundi, Lates calcarifer, is an important tropical aquaculture species that undergo natural and permanent male to female sex change, a process for which the exact underlying molecular mechanisms are still unknown. To elucidate whether DNA methylation is involved in sex control of barramundi, a next-generation bisulfite amplicon sequencing approach was used to target 146 CpG sites within proximal promoters and first exons of seven sex-related genes (dmrt1, cyp19a1, amh, foxl2, nr5a2, sox8 and sox9) of 24 testis and 18 ovaries of captive and wild adult barramundi. Moreover, comparative expression profiles of the key dmrt1 and cyp19a1 genes were further investigated using RT-qPCR and Sanger sequencing approaches, whereas expression levels of remaining targeted genes were based on available literature for the species. Results showed that cyp19a1 and amh were more methylated in males, whereas dmrt1 and nr5a2 were more methylated in females (P < 0.001), with no gender differences found for foxl2, sox8 or sox9 genes (P > 0.05). Sex-biased promoter DNA methylation was inversely related to gene expression only for dmrt1 and nr5a2, and directly related to amh expression, whereas no differences in cyp19a1 expression were found between testes and ovaries. Notably, unique sex-specific alternative splicing of dmrt1 and cyp19a1 were discovered, whereby males lacked the full-length aromatase coding cyp19a1 mRNA due to partial or total exon splicing, and females lacked the dmrt1 exon containing the DM-domain sequence. This study advances the current knowledge aiming to elucidate the genetic mechanisms within male and female gonads of this large protandrous hermaphrodite by providing the first evidence of epigenetics and alternative splicing simultaneously affecting key genes (cyp19a1 and dmrt1) central to sex differentiation pathways.

Approximately 2,500 weights and corresponding images of harvested Lates calcarifer (Asian seabass or barramundi) were collected at three different locations in Queensland, Australia. Two instances of the LinkNet-34 segmentation Convolutional Neural Network (CNN) were trained. The first one was trained on 200 manually segmented fish masks with excluded fins and tails. The second was trained on 100 whole-fish masks. The two CNNs were applied to the rest of the images and yielded automatically segmented masks. The one-factor and two-factor simple mathematical weight-from-area models were fitted on 1072 area-weight pairs from the first two locations, where area values were extracted from the automatically segmented masks. When applied to 1,400 test images (from the third location), the one-factor whole-fish mask model achieved the best mean absolute percentage error (MAPE), MAPE = 4.36%. Direct weight-from-image regression CNNs were also trained, where the no-fins based CNN performed best on the test images with MAPE = 4.28%.