Introduction
The global catfish industry, estimated to be worth over USD 7 billion annually, faces a critical "productivity plateau" driven by unresolved reproductive barriers and climate volatility. While Clarias gariepinus × C. macrocephalus hybrids dominate production, total male hybrid sterility and declining broodstock quality under thermal stress threaten long-term food security. This work presents a paradigm shift: utilizing the first haplotype-resolved, chromosome-scale genome of C. macrocephalus to unlock precision breeding through Genome-Wide Association Studies (GWAS) and AI-based hatchery management.
Materials and Methods
We established an 880 Mb reference genome using PacBio HiFi and Hi-C scaffolding. This genomic blueprint was integrated with: 1) Cytogenetic mapping of meiotic failure in hybrids; 2) GWAS and Genomic Selection (GS) protocols to identify markers for growth and disease resistance; 3) Deep learning (YOLOv11/ResNet18) for automated cranial-based species classification; and 4) Bio-economic modeling of thermal impacts on larval yolk sac absorption efficiency (mg kg���1).
Results
The chromosome-scale assembly (27 pseudo-chromosomes, N50: 33.48 Mb) revealed that satellite DNA divergence (CLA-SAT-149) is the primary driver of meiotic asynapsis, explaining the F1��� hybrid sterility bottleneck. GWAS facilitated the identification of loci associated with the high-growth shooter phenotype, providing a definitive roadmap for genomic selection. Furthermore, a dedicated genome browser with multiple Clarias individuals was developed to enable breeders and researchers to perform comparative genomic analysis, execute population-level GWAS, and plan data-driven breeding programs. For industry application, we identified the black testes phenotype as a major cause of hatchery infertility and demonstrated that elevated temperatures accelerate yolk sac depletion, reducing the larval safety window. Our AI framework provided a 99.5% accurate, non-invasive tool for identifying purebred spawners to mitigate stock contamination.
Discussion
Transitioning C. macrocephalus from a genomic orphan to a genomic model enables the first implementation of marker-assisted and genomic selection in Clarias aquaculture. This integrative framework solves the sterility puzzle, provides AI tools for field-level quality control, and addresses the global fish supply gap through a sustainable, climate-resilient One Health approach.
Acknowledgment
This study was supported by the Program Management Unit for Human Resources and Institutional Development, Research and Innovation (PMU-B) under Contract No. B13F670053; the Program Management Unit for Competitiveness (PMU-C), through the Global Partnership Program, under Contract No. C23F670224; the National Research Council of Thailand (NRCT) through the High-Potential Research Team Grant Program under Contract No. N42A660605; Betagro Group under Grant No. 6501.0901.1/68; the Kasetsart University Research and Development Institute under Grant No. FF(KU)61.69; and the International SciKU Branding (ISB) initiative of the Faculty of Science, Kasetsart University.
References
Andres, Q. L. S., Singchat, W., Panthum, T., Muangmai, N., Duengkae, P., Prasanpan, J., Srisapoome, P., Matsuda, Y., & Srikulnath, K. (2026). Chromosome-scale, haplotype-resolved genome of bighead catfish for conservation and aquaculture. Scientific Data, 13, Article 8. https://doi.org/10.1038/s41597-025-06325-6
Lisachov, A., Panthum, T., Singchat, W., Prasanpan, J., Griffin, D., Matsuda, Y., & Srikulnath, K. (2026). Mixed modes of satellite DNA and transposable element evolution, which shape the repetitive landscape of air-breathing catfishes (Clarias). Chromosome Research, 34, Article 7. https://doi.org/10.1007/s10577-026-09797-3
Treeprapin, K., Bandatang, R., Panthum, T., Singchat, W., Prasanpan, J., Srikulnath, K., & Trirongjitmoah, S. (2025). Enhancing clariid catfish species classification: A deep learning framework utilizing cranial morphology and explainable AI. Smart Agricultural Technology, 12, Article 101165. https://doi.org/10.1016/j.atech.2025.101165