Developing a parsimonius predictor for binary traits in sugar beet (Beta vulgaris)

Filippo Biscarini, Simone Marini, Piergiorgio Stevanato, Chiara Broccanello, Riccardo Bellazzi, Nelson Nazzicari

Research output: Contribution to journalArticle

6 Citations (Scopus)

Abstract

Binary traits are often encountered in plant (and animal) breeding. These include resistance/susceptibility to diseases or the presence/absence of a given characteristic. Root vigor in sugar beet is related to nutrient uptake from the soil and sugar yield and can be classified as either high or low, thus providing an example of binary trait of agronomic importance. Genomic data may be used for early prediction of seedlings root vigor in sugar beet breeding programmes. In this context, it may be of theoretical and practical interest to determine the minimum set of data needed for accurate predictions. A panel of 175 SNP markers was used to genotype 123 sugar beet individual plants. SNPs were ranked based on their predictive ability in a model selection algorithm. Starting from the bottom (least relevant SNP), one SNP at a time was removed and the predictive ability of the remaining SNPs assessed. The accuracy of prediction was in general very high, close to 100 %. Only starting from ≤30 SNPs in the model, the prediction accuracy became less stable and began to decrease. Based on results, a set of 30–50 SNPs can be recommended for accurate prediction of root vigor in sugar beet populations. The described procedure is in principle applicable to any binary trait in any plant (or animal) species of agricultural interest.

Original languageEnglish
JournalMolecular Breeding
Volume35
Issue number1
DOIs
Publication statusPublished - 2015

Fingerprint

Beta vulgaris
sugar beet
Single Nucleotide Polymorphism
prediction
vigor
animal breeding
plant breeding
agronomic traits
nutrient uptake
Disease Susceptibility
Seedlings
sugars
genomics
Breeding
seedlings
genotype
Soil
breeding
Genotype
soil

Keywords

  • Binary traits
  • Dataset reduction
  • Genomic predictions
  • Parsimonious predictor
  • SNP
  • Sugar beet

ASJC Scopus subject areas

  • Plant Science
  • Molecular Biology
  • Agronomy and Crop Science
  • Genetics
  • Biotechnology

Cite this

Developing a parsimonius predictor for binary traits in sugar beet (Beta vulgaris). / Biscarini, Filippo; Marini, Simone; Stevanato, Piergiorgio; Broccanello, Chiara; Bellazzi, Riccardo; Nazzicari, Nelson.

In: Molecular Breeding, Vol. 35, No. 1, 2015.

Research output: Contribution to journalArticle

Biscarini, Filippo ; Marini, Simone ; Stevanato, Piergiorgio ; Broccanello, Chiara ; Bellazzi, Riccardo ; Nazzicari, Nelson. / Developing a parsimonius predictor for binary traits in sugar beet (Beta vulgaris). In: Molecular Breeding. 2015 ; Vol. 35, No. 1.
@article{7f80775a09014e7bb1fa41f2c6140fd8,
title = "Developing a parsimonius predictor for binary traits in sugar beet (Beta vulgaris)",
abstract = "Binary traits are often encountered in plant (and animal) breeding. These include resistance/susceptibility to diseases or the presence/absence of a given characteristic. Root vigor in sugar beet is related to nutrient uptake from the soil and sugar yield and can be classified as either high or low, thus providing an example of binary trait of agronomic importance. Genomic data may be used for early prediction of seedlings root vigor in sugar beet breeding programmes. In this context, it may be of theoretical and practical interest to determine the minimum set of data needed for accurate predictions. A panel of 175 SNP markers was used to genotype 123 sugar beet individual plants. SNPs were ranked based on their predictive ability in a model selection algorithm. Starting from the bottom (least relevant SNP), one SNP at a time was removed and the predictive ability of the remaining SNPs assessed. The accuracy of prediction was in general very high, close to 100 {\%}. Only starting from ≤30 SNPs in the model, the prediction accuracy became less stable and began to decrease. Based on results, a set of 30–50 SNPs can be recommended for accurate prediction of root vigor in sugar beet populations. The described procedure is in principle applicable to any binary trait in any plant (or animal) species of agricultural interest.",
keywords = "Binary traits, Dataset reduction, Genomic predictions, Parsimonious predictor, SNP, Sugar beet",
author = "Filippo Biscarini and Simone Marini and Piergiorgio Stevanato and Chiara Broccanello and Riccardo Bellazzi and Nelson Nazzicari",
year = "2015",
doi = "10.1007/s11032-015-0197-5",
language = "English",
volume = "35",
journal = "Molecular Breeding",
issn = "1380-3743",
publisher = "Springer Netherlands",
number = "1",

}

TY - JOUR

T1 - Developing a parsimonius predictor for binary traits in sugar beet (Beta vulgaris)

AU - Biscarini, Filippo

AU - Marini, Simone

AU - Stevanato, Piergiorgio

AU - Broccanello, Chiara

AU - Bellazzi, Riccardo

AU - Nazzicari, Nelson

PY - 2015

Y1 - 2015

N2 - Binary traits are often encountered in plant (and animal) breeding. These include resistance/susceptibility to diseases or the presence/absence of a given characteristic. Root vigor in sugar beet is related to nutrient uptake from the soil and sugar yield and can be classified as either high or low, thus providing an example of binary trait of agronomic importance. Genomic data may be used for early prediction of seedlings root vigor in sugar beet breeding programmes. In this context, it may be of theoretical and practical interest to determine the minimum set of data needed for accurate predictions. A panel of 175 SNP markers was used to genotype 123 sugar beet individual plants. SNPs were ranked based on their predictive ability in a model selection algorithm. Starting from the bottom (least relevant SNP), one SNP at a time was removed and the predictive ability of the remaining SNPs assessed. The accuracy of prediction was in general very high, close to 100 %. Only starting from ≤30 SNPs in the model, the prediction accuracy became less stable and began to decrease. Based on results, a set of 30–50 SNPs can be recommended for accurate prediction of root vigor in sugar beet populations. The described procedure is in principle applicable to any binary trait in any plant (or animal) species of agricultural interest.

AB - Binary traits are often encountered in plant (and animal) breeding. These include resistance/susceptibility to diseases or the presence/absence of a given characteristic. Root vigor in sugar beet is related to nutrient uptake from the soil and sugar yield and can be classified as either high or low, thus providing an example of binary trait of agronomic importance. Genomic data may be used for early prediction of seedlings root vigor in sugar beet breeding programmes. In this context, it may be of theoretical and practical interest to determine the minimum set of data needed for accurate predictions. A panel of 175 SNP markers was used to genotype 123 sugar beet individual plants. SNPs were ranked based on their predictive ability in a model selection algorithm. Starting from the bottom (least relevant SNP), one SNP at a time was removed and the predictive ability of the remaining SNPs assessed. The accuracy of prediction was in general very high, close to 100 %. Only starting from ≤30 SNPs in the model, the prediction accuracy became less stable and began to decrease. Based on results, a set of 30–50 SNPs can be recommended for accurate prediction of root vigor in sugar beet populations. The described procedure is in principle applicable to any binary trait in any plant (or animal) species of agricultural interest.

KW - Binary traits

KW - Dataset reduction

KW - Genomic predictions

KW - Parsimonious predictor

KW - SNP

KW - Sugar beet

UR - http://www.scopus.com/inward/record.url?scp=84921299305&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84921299305&partnerID=8YFLogxK

U2 - 10.1007/s11032-015-0197-5

DO - 10.1007/s11032-015-0197-5

M3 - Article

AN - SCOPUS:84921299305

VL - 35

JO - Molecular Breeding

JF - Molecular Breeding

SN - 1380-3743

IS - 1

ER -