A Genome-Wide Association Study Identifies Multiple Regions Associated with Head Size in Catfish

Xin Geng, Shikai Liu, Jun Yao, Lisui Bao, Jiaren Zhang, Chao Li, Ruijia Wang, Jin Sha, Peng Zeng, Degui Zhi, Zhanjiang LiuG3 2016 · DOI

TL;DR

This study leverages a genome-wide association study (GWAS) in catfish to pinpoint genetic regions influencing head size, a trait crucial for both evolutionary adaptation and aquaculture yield. They identified significant genetic loci associated with head length and width, uncovering candidate genes involved in bone development that show evolutionary conservation across diverse species.

What Problem Does This Paper Address?

Skull morphology is a fundamental aspect of species adaptation to their environments. In aquaculture, head size is particularly important because it directly impacts fillet yield, a key economic driver. Despite this, the underlying genetic mechanisms controlling head size in commercially important fish species remain largely unknown. Catfish, a major aquaculture species in the US, presents an ideal model for this research, especially given the availability of genomic resources and the economic incentive to improve fillet yield through selective breeding. This paper aims to bridge this knowledge gap by identifying quantitative trait loci (QTL) associated with head size in catfish.

Methods

The researchers employed a genome-wide association study (GWAS) using a high-density 250K SNP array in a backcross hybrid catfish population. To handle the complex genetic structure of the population and account for phenotypic variation due to body weight, they used a two-step analytical approach. First, phenotypic data were adjusted for body weight using linear regression. Second, GWAS was performed using two methods: EMMAX (Efficient Mixed-Model Association eXpedited) and QFAM (family-based association test for quantitative traits). QFAM was found to provide more statistical power for this family-based dataset. Candidate genes within the identified QTL regions were then examined for their known functions related to bone development and skull morphology.

Datasets

The study utilized a dataset of 556 backcross hybrid catfish individuals. Phenotypic data included measurements for head length, head width, and head depth, along with body weight. Genotypic data were obtained using a catfish 250K SNP array, resulting in the analysis of 218,784 SNPs after quality control. The genetic marker map was constructed based on the channel catfish genome sequence.

Key Insights and Ideas

  • Head size is genetically influenced: The study successfully identified specific genomic regions (linkage groups) associated with variations in head length and width, demonstrating a clear genetic basis for these traits.
  • Evolutionary conservation of skull morphology genes: Candidate genes identified in catfish for head size control are homologous to genes known to be involved in skull morphology in a wide range of species, from amphibians to mammals. This suggests that the genetic pathways regulating skull development are ancient and conserved across vertebrates.
  • Functional hubs of candidate genes: The research observed enrichment of genes within the same pathway (e.g., small GTPases) in associated QTL regions, suggesting these genes might work together in "functional hubs" to regulate head size.
  • QFAM outperforms EMMAX for family-based GWAS: For this specific family-based population, the QFAM method demonstrated superior statistical power compared to EMMAX, highlighting the importance of choosing appropriate analytical tools for different population structures.

Key Results

  • Head Length: A significantly associated region was identified on linkage group (LG) 7 for head length. Additionally, LGs 7, 9, and 16 contained suggestively associated regions.
  • Head Width: Significantly associated regions were found on LG9, with suggestive associations on LGs 5 and 7.
  • Head Depth: No significant or suggestively associated regions were identified for head depth.
  • Variance Explained: The most significant SNP on LG7 for head length explained 0.02% of the phenotypic variance, with other suggestively associated regions explaining an additional 0.05%. For head width, associated QTL explained a total of 0.08% of the phenotypic variance.
  • Candidate Genes: Identified candidate genes within the QTL regions are involved in bone development, including small GTPases, sphingomyelin phosphodiesterase, fibroblast growth factor-binding protein, and regulators of G-protein signaling.

Limitations and Future Work

The authors acknowledge that the identified QTL regions are relatively large (±1 Mb), which is partly due to the limited recombination in backcross progenies and potential issues with genome assembly. This necessitates further fine-mapping to pinpoint the exact causative genes. They also note that while homologs of candidate genes are conserved, it's not necessarily the orthologous genes themselves that are responsible for variations in head shape across species, but rather specific variants within these genes. Future work will focus on fine-mapping these QTL and utilizing this information for marker-assisted selection (MAS) to improve catfish breeding for desirable head size traits, ultimately enhancing aquaculture production.

Summary updated

Abstract

Skull morphology is fundamental to evolution and the biological adaptation of species to their environments. With aquaculture fish species, head size is also important for economic reasons because it has a direct impact on fillet yield. However, little is known about the underlying genetic basis of head size. Catfish is the primary aquaculture species in the United States. In this study, we performed a genome-wide association study using the catfish 250K SNP array with backcross hybrid catfish to map the QTL for head size (head length, head width, and head depth). One significantly associated region on linkage group (LG) 7 was identified for head length. In addition, LGs 7, 9, and 16 contain suggestively associated regions for head length. For head width, significantly associated regions were found on LG9, and additional suggestively associated regions were identified on LGs 5 and 7. No region was found associated with head depth. Head size genetic loci were mapped in catfish to genomic regions with candidate genes involved in bone development. Comparative analysis indicated that homologs of several candidate genes are also involved in skull morphology in various other species ranging from amphibian to mammalian species, suggesting possible evolutionary conservation of those genes in the control of skull morphologies.
BibTeX
@inproceedings{geng2016genome,
  author = {Xin Geng and Shikai Liu and Jun Yao and Lisui Bao and Jiaren Zhang and Chao Li and Ruijia Wang and Jin Sha and Peng Zeng and Degui Zhi and Zhanjiang Liu},
  title = {{A Genome-Wide Association Study Identifies Multiple Regions Associated with Head Size in Catfish}},
  booktitle = {G3},
  year = {2016},
  doi = {10.1534/g3.116.032201},
}
Figure