Sign in Subscribe
Machine learning

Genetic Basis of Growth Traits in Snapper Identified

Strategic Insights

— 1 min read

Researchers at a leading institution have made a breakthrough in understanding the genetic basis of growth-related traits in snapper (Chrysophrys auratus), a species commonly farmed for food. The study, published on biorxiv.org, involves the analysis of phenotypic and genomic data from a selectively bred population of snapper. The researchers used a high-throughput, image-based phenotyping pipeline to extract 13 measurements, which were then correlated with manually measured weight and fork length. Heritabilities were estimated for each trait, and genome-wide association studies (GWAS) were performed to identify growth-associated SNPs. The team also implemented machine learning (ML) models in XGBoost to predict growth traits based on SNP genotypes.

Key Takeaways:

  • The study analyzed phenotypic and genomic data from a selectively bred population of snapper to identify genetic variants associated with key growth traits.
  • The high-throughput, image-based phenotyping pipeline extracted 13 measurements, which were then correlated with manually measured weight and fork length.
  • The researchers estimated heritabilities for each trait and performed GWAS to identify 24 SNPs significantly associated with growth traits.
  • The top ML growth SNPs showed some congruence with the GWAS growth SNPs, with 75% of the GWAS SNPs used by the ML model to predict weight.
  • The study highlights the utility of integrating computer vision-based phenotyping with GWAS and ML for trait prediction in aquaculture species.
  • The findings contribute to the development of genomic selection tools for snapper breeding and provide insights into potential biological mechanisms underlying growth variation.
  • The study focuses on the selective breeding of snapper, which is a species commonly farmed for food.

Statistics:

  • 13 phenotypic measurements were extracted using a high-throughput, image-based phenotyping pipeline.
  • 24 SNPs were identified as significantly associated with growth traits via GWAS.
  • The ML approach achieved moderate levels of predictability, with the top growth SNPs showing some congruence with the GWAS growth SNPs.
  • 75% of the GWAS SNPs were used by the ML model to predict weight.
  • Genome-wide association studies (GWAS) were performed to identify growth-associated SNPs.

Sources:

  • biorxiv.org/content/10.1101/2025.05.29.656727v1

Read more