Adult gonadal transcriptomes reflect tissue specialization and reproductive activity. We compared testes and ovaries from seven male and seven female mud carp (Cirrhinus molitorella) and evaluated the results in an independent six-fish public cohort and by RT-qPCR in a further 12 fish. DESeq2 analysis identified 12,757 sex-biased loci among 19,322 tested loci, comprising 8213 testis-biased and 4544 ovary-biased loci. Testis-biased loci were enriched for cilium and axoneme assembly, whereas ovary-biased loci were enriched for RNA processing and ribosome biogenesis. In a common-reference comparison with the public cohort, 9045 of 9104 shared differentially expressed loci (99.4%) had concordant directions, and fold changes across 17,283 loci were strongly correlated (Pearson r = 0.917). RT-qPCR of seven candidate genes reproduced the RNA-seq direction in all cases, six of them significantly, and RT-qPCR and RNA-seq fold changes were closely correlated (r = 0.984). Protein phylogeny and local synteny reassigned two strongly testis-biased loci originally annotated as foxl2 and wnt4a to foxl2l-like and wnt4b-like paralogues. Differential splicing analysis identified 923 skipped-exon and mutually exclusive-exon events across 743 loci. These results provide a validated set of sex-biased genes and refined paralogue annotations for future studies of reproduction and sex control in mud carp.
Body weight is the principal economic trait in snakehead aquaculture, yet the genetic architecture of growth in the northern snakehead (Channa argus) remains largely unexplored. We performed a genome-wide association study (GWAS) for body weight in 205 cultured C. argus genotyped with a 50K SNP array. After quality control, 47,123 autosomal SNPs were tested under a univariate linear mixed model in GEMMA that accounted for genomic relatedness. One SNP on chromosome 7 (Chr07:1,329,732) exceeded the Bonferroni genome-wide threshold, and 14 SNPs reached a suggestive threshold, forming two candidate clusters on Chr07 and Chr24. Annotating candidate intervals (±100 kb) against the C. argus genome assembly yielded 93 candidate genes. Hypergeometric enrichment identified 44 Gene Ontology (GO) terms and 4 KEGG pathways, dominated by endolysosomal lumen, serine-type endopeptidase activity, collagen/extracellular-matrix catabolism, bone resorption, and the apoptosis, lysosome, phagocytosis and antigen-processing pathways. The single genome-wide significant SNP fell in an intron of a collagen alpha-1(II) chain gene (COL2A1/zgc:113232), consistent with the collagen/extracellular-matrix enrichment signal; other plausible growth-related candidates included CTSS, S100A4, PYGM, DLAT, MSX2 and ZNRF3. The functional signal was largely driven by a tandem protease/cathepsin gene cluster on Chr07, indicating a regional rather than a genome-wide pathway effect. The lead SNP genotype was associated with a ~240 g (≈30%) difference in body weight, and this effect was reproduced in an independent cohort of 200 fish. These findings provide the first GWAS-based candidate loci for body weight in C. argus and a promising marker for future marker-assisted breeding, though validation to date is limited to a single additional cohort from the same broodstock source; broader multi-population testing will be needed before routine application.