deepGBLUP: joint deep learning networks and GBLUP framework for accurate genomic prediction of complex traits in Korean native cattle.

Journal: Genetics, selection, evolution : GSE
PMID:

Abstract

BACKGROUND: Genomic prediction has become widespread as a valuable tool to estimate genetic merit in animal and plant breeding. Here we develop a novel genomic prediction algorithm, called deepGBLUP, which integrates deep learning networks and a genomic best linear unbiased prediction (GBLUP) framework. The deep learning networks assign marker effects using locally-connected layers and subsequently use them to estimate an initial genomic value through fully-connected layers. The GBLUP framework estimates three genomic values (additive, dominance, and epistasis) by leveraging respective genetic relationship matrices. Finally, deepGBLUP predicts a final genomic value by summing all the estimated genomic values.

Authors

  • Hyo-Jun Lee
    Department of Bio-AI Convergence, Chungnam National University, 305-764, Daejeon, Korea.
  • Jun Heon Lee
    Division of Animal and Dairy Science, Chungnam National University, 305-764, Daejeon, Korea.
  • Cedric Gondro
    CSIRO Data61, Canberra, Australian Capital Territory, Australia; College of Agriculture and Natural Resources, Michigan State University, East Lansing, MI, USA.
  • Yeong Jun Koh
    Department of Computer Science and Engineering, Chungnam National University, 305-764, Daejeon, Korea. yjkoh@cnu.ac.kr.
  • Seung Hwan Lee
    Division of Animal and Dairy Science, Chungnam National University, 305-764, Daejeon, Korea. slee46@cnu.ac.kr.