Sex identification in rainbow trout using genomic information and machine learning
2024

Identifying the Sex of Rainbow Trout Using Genomics and Machine Learning

Sample size: 1362 publication Evidence: high

Author Information

Author(s): Kudinov Andrei A., Kause Antti

Primary Institution: Natural Resources Institute Finland

Hypothesis

Can machine learning accurately predict the sex of rainbow trout using genomic data?

Conclusion

The study demonstrated that the Extreme Gradient Boosting method can accurately predict the sex of rainbow trout with a prediction accuracy of 98%.

Supporting Evidence

  • The method achieved a prediction accuracy of 98% in real data from the Finnish breeding program.
  • Simulated datasets showed accuracies of 1.0 and 0.60 for low and high genotyping error rates, respectively.
  • The XGB model was robust to cases where a larger part of the data was masked.

Takeaway

Scientists used a computer program to figure out if rainbow trout are boys or girls by looking at their genes, and it worked really well!

Methodology

The study used the Extreme Gradient Boosting approach to predict sex from genomic data without prior knowledge of suitable markers.

Potential Biases

Potential bias in sex identification due to reliance on phenotypic data that may be inaccurately recorded.

Limitations

The accuracy of the method may vary with different populations and the model may need retraining with new data.

Participant Demographics

The study involved 1362 rainbow trout from the Finnish Rainbow Trout Breeding Program, including 491 males and 871 females.

Digital Object Identifier (DOI)

10.1186/s12711-024-00944-0

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication