Article banner region

Application of Genome Wide SNPs in Single Step Genomic Analysis

  • Photo: 

NOVA PhD course of 1.5 ECTS, organised by Sreten Andonov, PhD, Swedish University of Agricultural Sciences.
Course dates and location: 2-6 September 2019, in Uppsala, Sweden.

Please note that this is a competence-enhancing course for PhD students and other scientists.

Course Description
This is a course on implementing genomic selection, where polygenic prediction using genome-wide SNPs provides high prediction accuracy for complex traits. The course will be given by prof Ignacy Misztal, PhD and Daniela Lourenco, PhD, from UGA, USA, and Sreten Andonov, PhD, SLU.

The course will include sections on: a) BLUPF90 family of programs: animal and maternal model, multiple trait and genomic model; b) Introduction to genomic selection, basis of SNP data, simulation of genomic data, data manipulation and bash scripting in Linux; c) Methods based on SNP estimation (SNP_BLUP, BayesX), methods based on genomic relationships, creation and handling of genomic relationship matrices with preGSf90, GBLUP, GREML and GGIBBBS; d) Theory of single-step GBLUP, single-step for populations under selection: bias, inflation, accuracy, forming single-step equations, quality control for G, validation methods, predictability for animals with records, cross-validation when few genotyped animals with records e.g. mortality or disease resistance; e) Estimating SNP effects for GBLUP-based methods, weighted GBLUP and ssGBLUP, linear weights, nonlinear A weights, genome-wide association (GWA) and p-values with postGSf90, experiences and future with ssGBLUP. Exercises: use of programs for data sets with single and multiple traits; data simulation and manipulation; use of programs with simulated data; single-step with simulated data set; application of weighted ssGBLUP and GWA. Literature will be delivered before the course.

Programme Outline
The course is characterized by a very strong connection to the breeding programs and population investigations. The course will last for 5 days. Each day there will be 5 hours of lectures: theoretical, demonstrations and case studies, followed by 3 hours computer exercises. From the exercises, each participant will learn how to apply the method explained before in real analysis. Additionally, it is expected each attendant will spend 1-hour on independent work after the classes. A short explanatory course (offered also online) will be given to participants a week before in order to set up their laptops to run programs on a Linux server.

Learning Outcomes
The course aims to build a strong understanding in application of methods of genomic selection. From theoretical developments to efficient programming and novel traits, students/attendants can obtain knowledge and experience necessary to achieve their career goals in academia or the industry. Students will be able to learn the latest knowledge from the world leading researchers/teachers in the area. The exercises are designed for students to be able to build skills for independent data preparation and evaluation. At the end of the course, they will be able to build a personal portfolio in use of statistical methods to use in research work and in commitment of genomics in breeding programs. PhD students can build a network not only within NOVA countries but also an international network.

Evaluation Elements
Independent completion of course exercises.

Pedagogical Approach
Theoretical lectures with many examples, and exercises with predefined assignments. In the exercises, assistance will be offered but strongly encouraging independent work, and there will be problem based homework.

Estimated Workload

  • 2 hours preparation (Linux computer setup)
  • 15 hours computer exercises
  • 25 hours lecture
  • 5 hours homework

Prerequisite Knowledge
At least MSc in Animal/Plant Science or participant in a residency program in veterinary science, animal/plant breeding or animal/plant genetics. Participants are expected to have understanding of animal /plant breeding and statistics, and genetic evaluation, and to be familiar with mixed model equations and quantitative genetics. Familiarity with Linux / Unix environments is expected for lab exercises. Each participant should bring his/her own laptop.

Admission for NOVA courses is handled by the course organiser/ the NOVA member institution organising the course. Please see the links in the margin for more information.

Apply here


Course fee will be charged (in accordance with the NOVA regulations)

  • PhD students - free
  • NOVA/BOVA teachers/researchers - 200 Euros + VAT
  • Participants from the industry - 400 Euros + VAT
Published 11. April 2019 - 10:22 - Updated 29. July 2019 - 13:23