Mots-Clés

Statistics, quantitative genetics, Genome Estimated Breeding Value

Description

Internship global context: The global population is estimated to reach approximately nine billion people by 2050, thus the demand for animal protein is expected to increase by 76% [1]. Such an increase questions the sustainability of our conventional food and feed production systems. At the same time, we also need to reduce the impact of agriculture on our environment [2]. Today, insect production is considered a sustainable alternative for food and feed production for several reasons. First, the suitable nutritional composition of edible insects [3] and second, the relatively low environmental impact its production involves compared to other conventional livestock production systems. However, non-conventional animal breeding raises the problem of adapted statistical and computational methods for genomic breeding value prediction due to the nature of the pooled genomic data and phenotypes estimates from different individuals while classical approaches use distinct genomic data and phenotype per each individual.

Internship tasks:

1. Bibliography on state-of-the-art statistical methods used in Genome Estimated Index Value (GEBV) such as genomic best linear unbiased predictor (GBLUP) [4], ridge regression BLUP (rrBLUP) [5], Bayesian LASSO (BL) [6], and reproducing kernel Hilbert space (RKHS) regression [7].

2. Testing of statistical approaches for GEBV with allele frequencies inferred from pooled sequencing data (Pool-seq).

3. Implementation of the algorithms in R

4. Performance evaluation of the proposed algorithms on simulated and real data sets. Hosting laboratory: The six months internship will takes place in the “Statistiques et Génome” team of the Laboratory of Mathematics and Modelisation of Evry (LaMME) in Evry, France http://www.math-evry.cnrs.fr/sg/welcome.

Supervision: The internship will be supervised by Pr. Christophe Ambroise, Professor of statistics at Paris-Saclay University (https://cambroise.github.io/) and Dr. Amin Madoui, CEA researcher in genomics in Fontenay-aux-Roses (https://madoui.github.io/)

Student profile: Master student in statistics with programming knowledge in R

References:

[1] Alexandratos N, Bruinsma J: World agriculture towards 2030/2050: the 2012 revision. 2012

[2] Steinfeld H, Gerber P, Wassenaar T, et al.: Livestock’s long shadow. 2006.

[3] Nowak V, Persijn D, Rittenschober D, et al.: Review of food composition data for edible insects. Food Chem. 2016; 193: 39–46.

[4] Goddard ME, Hayes BJ, Meuwissen THE. Using the genomic relationship matrix to predict the accuracy of genomic selection. J Anim Breed Genet. 2011;128(6):409–421. doi: 10.1111/j.1439-0388.2011.00964.x

[5] Meuwissen THE, Hayes BJ, Goddard ME. Prediction of total genetic value using genome-wide dense marker maps. Genetics. 2001;157(4):1819–1829.

[6] Legarra A, Robert-Granie C, Croiseau P, Guillaume F, Fritz S. Improved Lasso for genomic selection. Genet Res. 2011;93(1):77–87. doi: 10.1017/S0016672310000534.