Learning from data: Plant breeding applications of machine learning

<p>Increasingly, new sources of data are being incorporated into plant breeding pipelines. Enormous amounts of data from field phenomics and genotyping technologies places data mining and analysis into a completely different level that is challenging from practical and theoretical standpoints. Intelligent decision-making relies on our capability of extracting from data useful information that may help us to achieve our goals more efficiently. Many plant breeders, agronomists and geneticists perform analyses without knowing relevant underlying assumptions, strengths or pitfalls of the employed methods. The study endeavors to assess statistical learning properties and plant breeding applications of supervised and unsupervised machine learning techniques. A soybean nested association panel (<em>aka</em>. SoyNAM) was the base-population for experiments designed <em>in situ</em> and <em>in silico</em>. We used mixed models and Markov random fields to evaluate phenotypic-genotypic-environmental associations among traits and learning properties of genome-wide prediction methods. Alternative methods for analyses were proposed.</p>

Name thesis:degree_name

Doctor of Philosophy (PhD)

Level thesis:degree_level

Dissertation

Discipline thesis:degree_discipline

Agronomy

Year

2016

Author dc:creator

• Xavier, Alencar

Contributors dc:contributor

• Katy M. Rainey
• William M. Muir
• Shaun Casteel
• Bruce Craig
• Tobert Rocheford

• Biological sciences
• Bayesian methods
• Genomic prediction
• Machine learning
• Markov random fields
• Mixed models
• Soybean breeding
• Agriculture
• Biostatistics
• Plant Sciences

Repository record dc:identifier

https://docs.lib.purdue.edu/open_access_dissertations/883

OAI identifier oai:identifier

oai:docs.lib.purdue.edu:open_access_dissertations-2076