About this course

This course is aimed at giving students all the skills needed to work in the animal breeding and genetics industry, going from modern phenomics methods to advances quantitative genetic models. The course emphasizes real world analysis and problem solving on datasets provided in partnership by Norwegian breeding companies and organisations. This practical experience allows students to apply their skills to a complex, multidisciplinary problem, culminating in the delivery of a report which provides a breeding strategy or product proposal tailored to the students' chosen dataset.

This course equips students with the theoretical and practical skills to drive genetic innovation through the integration of the latest tools and models in phenomics, linear mixed models for breeding value estimation, including genomics selection. Starting with the generation of novel phenotypes, students will work with various digital data streams: such as images, sensor data, and vibrational spectroscopy. To develop new, precise measurements of traits critical to modern breeding programs. Through hands-on exercises using R, students will gain proficiency in data analysis, and interpretation for practical applications in precision breeding.

The course then transitions into advanced linear mixed modeling techniques, variance component estimation, and breeding value prediction to enable effective selection decisions. Students will explore various statistical models, including univariate animal models, multivariate models, and models incorporating genotype-environment interactions. Prediction errors and model comparisons will be covered to reinforce understanding and application of these statistical methods.

Students expand their linear mixed model understanding to state of the art genomic selection models, relationship matrices and genome wide association studies.

Learning outcome

Knowledge:

Students will gain a theoretical understanding of how to evaluate:

  • The merit of new phenotypes from different digital technologies.
  • The model fit of linear mixed models including BLUP, SNPBLUP, GBLUP, GWAS and multitrait, variance components
  • How traits can be included in a selection index for genetic gain

Skills:

  • Data analysis skills to include various data types like image/video, vibrational spectroscopy and sensors.
  • Construct mixed model equations for models with random and fixed effects
  • The abilities to estimate variance components and predict breeding values ​​​​with commercial software (ASReml)
  • Simulate a breeding program with the addition of new strategies
  • Contribute in a team to present a multidisciplinary project
  • Write, report, discuss and defend viewpoints in an individually produced report

Competence.

Students will be confident when exposed to new datasets and multidisciplinary breeding problems and have the ability to find optimal solutions for genetic gain and socially responsible animal breeding.

  • Learning activities

    • Weekly lectures will cover theoretical concepts behind the course modules
    • Weekly in person data labs will have focused data manipulation and analysis in R and other relevant software
    • Early presentation on your chosen dataset, your research question and your approach (Mandatory ungraded activity)
    • Final presentation to the class and instructors on your results and conclusion (Mandatory ungraded activity)
    • A final written report on the student's choice of given datasets towards the generation of genetic gain. (Graded activity).

  • Teaching support

    Teachers will provide active guidance during in-person lectures, data labs, and tutorials, with additional assistance available as needed. The course includes a mix of in-person and online lectures, discussions, presentations and individual studies. Learning is problem-oriented, with required assignments and a final report.

  • Syllabus

    -

  • Recommended prerequisites

    Basic programming skills in R or Python and an understanding of linear regression, multiple regression and analysis of variance. Basic understanding of quantitative genetics, plant or animal breeding.

  • Assessment method

    Internal censors will evaluate written reports. Internal censors and students will watch the presentations.

  • About use of AI

  • Examiner scheme

    Internal examiner assesses report quality.

  • Mandatory activity

    Two presentations will be given during the course. The first presentation is very short and the student describes their chosen dataset and their research question. Instructors will approve the dataset as appropriate or recommend chanding the dataset or the research question. The final presentation which the student presents to the class on the findings and recommendations. Peer and instructor questions are asked and feed back given. Students can use this feedback to help make any changes to their final written report.

  • Teaching hours

    Four hours per week of in person data tutorials and lectures. Weekly online lectures 2 hours per week.