Multi-trait, Multi-environment Deep Learning Modeling for Genomic-Enabled Prediction of Plant Traits

Abstract

Multi-trait and multi-environment data are common in animal and plant breeding programs. However, what is lacking are more powerful statistical models that can exploit the correlation between traits to improve prediction accuracy in the context of genomic selection (GS). Multi-trait models are more complex than univariate models and usually require more computational resources, but they are preferred because they can exploit the correlation between traits, which many times helps improve prediction accuracy. For this reason, in this paper we explore the power of multi-trait deep learning (MTDL) models in terms of prediction accuracy. The prediction performance of MTDL models was compared to the performance of the Bayesian multi-trait and multi-environment (BMTME) model proposed by Montesinos-López et al. (2016), which is a multi-trait version of the genomic best linear unbiased prediction (GBLUP) univariate model. Both models were evaluated with predictors with and without the genotype×environment interaction term. The prediction performance of both models was evaluated in terms of Pearson’s correlation using cross-validation. We found that the best predictions in two of the three data sets were found under the BMTME model, but in general the predictions of both models, BTMTE and MTDL, were similar. Among models without the genotype×environment interaction, the MTDL model was the best, while among models with genotype×environment interaction, the BMTME model was superior. These results indicate that the MTDL model is very competitive for performing predictions in the context of GS, with the important practical advantage that it requires less computational resources than the BMTME model.

Shown for reading. Its publisher allows quoting it, not republishing it.

References 39

1. 1.

   Alipanahi B. Delong A. Weirauch M. T. Frey B. J., 2015 Predicting the sequence specificities of DNA- and RNA-binding proteins by deep learning. Nat. Biotechnol. 33: 831–838.

   Not in our library10.1038/nbt.3300

2. 2.
   DeepCpG: accurate prediction of single-cell DNA methylation states using deep learning · 2017

   Corrected
3. 3.

   Bertan I. Carvalho F. I. F. Oliveira A. C. Silva J. A. G. Benin G., 2009 Efeitos da heterose e endogamia em caracteres de importância agronômica em trigo. Rev. Ceres 56: 753–763.

   Not in our library

4. 4.
   Accuracy of multi-trait genomic selection using different methods · 2011

   No notice
5. 5.

   Castro A. F. N. M. Castro R. V. O. Oliveira A. C. O. Lima J. E. Santos R. C., 2013 Multivariate analysis for the selection of eucalyptus clones destined for charcoal production. Pesqui. Agropecu. Bras. 48: 627–635.

   Not in our library10.1590/s0100-204x2013000600008

6. 6.

   Chollet F. Allaire J. J., 2017 Deep Learning with R. Manning Publications, Manning Early Access Program, Ed. 1st MEA, New Delhi, India.

   Not in our library

7. 7.

   Colleau J. J. Fritz S. Guillaume F. Baur A. Dupassieux D., 2009 Simulating the potential of genomic selection in dairy cattle breeding. Rencontres, Recherches. Ruminants 16: 419.

   Not in our library

8. 8.
   Genomic Prediction in Maize Breeding Populations with Genotyping-by-Sequencing · 2013

   No notice
9. 9.
   Genomic Prediction of Gene Bank Wheat Landraces · 2016

   No notice
10. 10.

    Everitt B. S. Dunn G., 1992 Applied Multivariate Data Analysis, Oxford University Press, New York, NY.

    Not in our library

11. 11.

    Fox, I., A. Lynn, J. Mamta, P.-B. Rodica, and W. Jenna, 2018 Deep Multi-Output Forecasting: Learning to Accurately Predict Blood Glucose Trajectories. In KDD ’18: The 24th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining, August 19–23, 2018, London, UK. ACM, New York, NY, USA

    Not in our library10.1145/3219819

12. 12.

    Gilmour A. Thompson R. Cullis B., 1995 Average Information REML: An Efficient Algorithm for Variance Parameter Estimation in Linear Mixed Models. Biometrics 51: 1440–1450.

    Not in our library10.2307/2533274

13. 13.

    Gulli A. Sujit P., 2017 Deep Learning with Keras: Implementing deep learning models and neural networks with the power of Python, Packt Publishing Ltd., Birmingham, UK.

    Not in our library

14. 14.

    He D. Kuhn D. Parida L., 2016 Novel applications of multitask learning and multiple output regression to multiple genetic trait prediction. Bioinformatics 32: i37–i43.

    Not in our library10.1093/bioinformatics/btw249

15. 15.

    Huang M. Chen L. Chen Z. 2015 Diallel analysis of combining ability and heterosis for yield and yield components in rice by using positive loci. Euphytica, 205: 3 7–50.

    Not in our library

16. 16.

    Jarquín D. Lemes da Silva C. Gaynor R. C. Poland J. Fritz A., 2017 Increasing Genomic-Enabled Prediction Accuracy by Modeling Genotype × Environment Interactions in Kansas Wheat. Plant Genome 10: 1–15.

    Not in our library10.3835/plantgenome2016.12.0130

17. 17.

    Jia Y. Jannink J.-L., 2012 Multiple-Trait Genomic Selection Methods Increase Genetic Value Prediction Accuracy. Genetics 192: 1513–1522.

    Not in our library10.1534/genetics.112.144246

18. 18.

    Jiang J. Zhang Q. Ma L. Li J. Wang Z., 2015 Joint prediction of multiple quantitative traits using a Bayesian multivariate antedependence model. Heredity 115: 29–36.

    Not in our library10.1038/hdy.2015.9

19. 19.
    Increased Prediction Accuracy in Wheat Breeding Trials Using a Marker × Environment Interaction Genomic Selection Model · 2015

    No notice
20. 20.
    Machine Learning Prediction of Cancer Cell Sensitivity to Drugs Based on Genomic and Chemical Properties · 2013

    No notice
21. 21.

    Montesinos-López O. A. Montesinos-López A. Crossa J. Toledo F. Pérez-Hernández O. 2016 A Genomic Bayesian Multi-trait and Multi-environment Model. G3: Genes|Genomes|Genetics (Bethesda), 6: 2725–2744.

    Not in our library10.1534/g3.116.032359

22. 22.
    Multi-environment Genomic Prediction of Plant Traits Using Deep Learners With Dense Architecture · 2018

    No notice
23. 23.

    Pérez P. de los Campos G., 2014 Genome-wide regression & prediction with the BGLR statistical package. Genetics 198: 483–495.

    Not in our library10.1534/genetics.114.164442

24. 24.
    A unified multitask architecture for predicting local protein properties. · 2012

    Not checked
25. 25.

    R Core Team, 2018. R: A language and environment for statistical computing. R Foundation for Statistical Computing. Vienna. Austria. ISBN 3–900051–07–0. URL http://www.R-project.org/.

    Not in our library

26. 26.
    Canopy Temperature and Vegetation Indices from High-Throughput Phenotyping Improve Accuracy of Pedigree and Genomic Selection for Grain Yield in Wheat · 2016

    No notice
27. 27.

    Schaeffer L. R., 2001 Multiple trait international bull comparisons. Livest. Prod. Sci. 69: 145–153.

    Not in our library10.1016/s0301-6226(00)00255-4

28. 28.

    Schulthess A. W. Zhao Y. Longin C. F. H. Reif J. C., 2017 Advantages and limitations of multiple-trait genomic prediction for Fusarium head blight severity in hybrid wheat (Triticum aestivum L.). Theor. Appl. Genet.

    Not in our library10.1007/s00122-017-3029-7

29. 29.

    Spencer M. Eickholt J. Cheng J. L., 2015 A Deep Learning Network Approach to ab initio Protein Secondary Structure Prediction. Ieee Acm T Comput Bi 12: 103–112.

    Not in our library10.1109/tcbb.2014.2343960

30. 30.

    Srivastava M. S. Khatri C. G., 1979 An Introduction to Multivariate Statistics. Elsevier North Holland, New York, NY, USA.

    Not in our library

31. 31.

    Srivastava N. Hinton G. Krizhevsky A. Sutskever I. Salakhutdinov R., 2014 Dropout: A Simple Way to Prevent Neural Networks from Overfitting. J. Mach. Learn. Res. 15: 1929–1958.

    Not in our library

32. 32.

    Sun J. Rutkoski J. Poland J. A. Crossa J. Jannink J. L., 2017 Multi-trait, random regression, or simply repeatability model in high-throughput phenotyping data improve genomic prediction for grain yield in wheat. Plant Genome 18: 1–12.

    Not in our library10.3835/plantgenome2016.11.0111

33. 33.

    Tavanaei A. Anandanadarajah N. Maida A. S. Loganantharaj R., 2017 A Deep Learning Model for Predicting Tumor Suppressor Genes and Oncogenes from PDB Structure. bioRxiv 1–10.

    Not in our library10.1101/177378

34. 34.

    VanRaden P. M., 2008 Efficient methods to compute genomic predictions. J. Dairy Sci. 91: 4414–4423.

    Not in our library10.3168/jds.2007-0980

35. 35.

    Wang Y. Wipf D. Ling Q. Chen W. Wassell I., 2015 Multitask learning for subspace segmentation. In ICML, 2015, pp. 1209–1217.

    Not in our library

36. 36.

    Xavier A. Hall B. Casteel S. Muir W. Rainey K. M., 2017 Using unsupervised learning techniques to assess interactions among complex traits in soybeans. Euphytica 213: 200

    Not in our library10.1007/s10681-017-1975-4

37. 37.

    Xiong T. Bao Y. Hu Z., 2014 Multiple-output support vector regression with a firefly algorithm for interval-valued stock price index forecasting. Knowl. Base. Syst. 55: 87–100.

    Not in our library10.1016/j.knosys.2013.10.012

38. 38.

    Yan Y. Ricci E. Subramanian G. Liu G. Lanz O., 2015 A multi-task learning framework for head pose estimation under target motion. IEEE Trans. Pattern Anal. Mach. Intell. 38: 1–14.

    Not in our library10.1109/tpami.2015.2477843

39. 39.

    Zhen X. Wang Z. Islam A. Bhaduri M. Chan I., 2015 Multi-scale deep networks and regression forests for direct biventricular volume estimation. Med. Image Anal. 30: 120–129.

    Not in our library10.1016/j.media.2015.07.003

Cited by papers we hold 4

• A review of deep learning applications for genomic selection · 2021
  No notice
• The Modern Plant Breeding Triangle: Optimizing the Use of Genomics, Phenomics, and Enviromics Data · 2021
  No notice
• Exploring Deep Learning for Complex Trait Genomic Prediction in Polyploid Outcrossing Species · 2020
  No notice
• Multi-environment Genomic Prediction of Plant Traits Using Deep Learners With Dense Architecture · 2018
  No notice

Reads

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.