|
|
| Acesso ao texto completo restrito à biblioteca da Embrapa Florestas. Para informações adicionais entre em contato com cnpf.biblioteca@embrapa.br. |
Registro Completo |
Biblioteca(s): |
Embrapa Florestas. |
Data corrente: |
17/11/2011 |
Data da última atualização: |
11/10/2017 |
Tipo da produção científica: |
Nota Técnica/Nota Científica |
Autoria: |
SILVA, F. F.; VARONA, L.; RESENDE, M. D. V. de; BUENO FILHO, J. S. S.; ROSA, G. J. M.; VIANA, J. M. S. |
Afiliação: |
Fabyano Fonseca Silva, UFV; Luis Varona, Universidad de Zaragoza; MARCOS DEON VILELA DE RESENDE, CNPF; Júlio Sílvio S. Bueno Filho, UFLA; Guilherme J. M. Rosa, University of Wisconsin; José Marcelo Soriano Viana, UFV. |
Título: |
A note on accuracy of Bayesian LASSO regression in GWS. |
Ano de publicação: |
2011 |
Fonte/Imprenta: |
Livestock Science, v. 142, p. 310-314, 2011. |
DOI: |
10.1016/j.livsci.2011.09.010 |
Idioma: |
Inglês |
Notas: |
Short communication. |
Conteúdo: |
Several genome wide selection (GWS) statistical methods have been proposed in the last years, and among these stands out the Bayesian LASSO (BL), which is a penalized regression method based on the regularization parameter (?) estimates. In general, the posterior mean values for ? are those that minimize the residual sum of squares (RSS) while controlling the L1 norm (absolute values) of the regression coefficients. However, another option is to use fixed values of ?, which is independent of this minimization process. Nevertheless, the most important aim of GWS is to make predictions about genomic breeding values (GBV=u) for individuals that have not been measured directly for the trait, and for this reason the parameter to maximize should be the accuracy (ru; ?u ). Thus, a question can arise as to whether such estimated ? values that minimize RSS are the same as that which maximize ru; ?u . In order to answer this question, this paper aims to provide methodological and computational resources in order to evaluate the influence of BL regularization parameter estimates on the correlation between true and estimated GBV (accuracy) depending on genetic structure of the target trait (few or many QTLs and low or medium heritability). In general, it is possible to report, on average, that GBV prediction is robust in relation to the ? estimation, since the different values for ? lead to similar accuracy values. Moreover, the fixed ? values grid request high computational costs, implying that the random ? method is more attractive, since it is much faster to use just one Gibbs sampler run, while the grid must to use one run for each fixed ? value. MenosSeveral genome wide selection (GWS) statistical methods have been proposed in the last years, and among these stands out the Bayesian LASSO (BL), which is a penalized regression method based on the regularization parameter (?) estimates. In general, the posterior mean values for ? are those that minimize the residual sum of squares (RSS) while controlling the L1 norm (absolute values) of the regression coefficients. However, another option is to use fixed values of ?, which is independent of this minimization process. Nevertheless, the most important aim of GWS is to make predictions about genomic breeding values (GBV=u) for individuals that have not been measured directly for the trait, and for this reason the parameter to maximize should be the accuracy (ru; ?u ). Thus, a question can arise as to whether such estimated ? values that minimize RSS are the same as that which maximize ru; ?u . In order to answer this question, this paper aims to provide methodological and computational resources in order to evaluate the influence of BL regularization parameter estimates on the correlation between true and estimated GBV (accuracy) depending on genetic structure of the target trait (few or many QTLs and low or medium heritability). In general, it is possible to report, on average, that GBV prediction is robust in relation to the ? estimation, since the different values for ? lead to similar accuracy values. Moreover, the fixed ? values grid request high computational costs, impl... Mostrar Tudo |
Palavras-Chave: |
Genome wide selection; Penalized regression; SNP markers. |
Categoria do assunto: |
-- |
Marc: |
LEADER 02366naa a2200241 a 4500 001 1906248 005 2017-10-11 008 2011 bl uuuu u00u1 u #d 024 7 $a10.1016/j.livsci.2011.09.010$2DOI 100 1 $aSILVA, F. F. 245 $aA note on accuracy of Bayesian LASSO regression in GWS.$h[electronic resource] 260 $c2011 500 $aShort communication. 520 $aSeveral genome wide selection (GWS) statistical methods have been proposed in the last years, and among these stands out the Bayesian LASSO (BL), which is a penalized regression method based on the regularization parameter (?) estimates. In general, the posterior mean values for ? are those that minimize the residual sum of squares (RSS) while controlling the L1 norm (absolute values) of the regression coefficients. However, another option is to use fixed values of ?, which is independent of this minimization process. Nevertheless, the most important aim of GWS is to make predictions about genomic breeding values (GBV=u) for individuals that have not been measured directly for the trait, and for this reason the parameter to maximize should be the accuracy (ru; ?u ). Thus, a question can arise as to whether such estimated ? values that minimize RSS are the same as that which maximize ru; ?u . In order to answer this question, this paper aims to provide methodological and computational resources in order to evaluate the influence of BL regularization parameter estimates on the correlation between true and estimated GBV (accuracy) depending on genetic structure of the target trait (few or many QTLs and low or medium heritability). In general, it is possible to report, on average, that GBV prediction is robust in relation to the ? estimation, since the different values for ? lead to similar accuracy values. Moreover, the fixed ? values grid request high computational costs, implying that the random ? method is more attractive, since it is much faster to use just one Gibbs sampler run, while the grid must to use one run for each fixed ? value. 653 $aGenome wide selection 653 $aPenalized regression 653 $aSNP markers 700 1 $aVARONA, L. 700 1 $aRESENDE, M. D. V. de 700 1 $aBUENO FILHO, J. S. S. 700 1 $aROSA, G. J. M. 700 1 $aVIANA, J. M. S. 773 $tLivestock Science$gv. 142, p. 310-314, 2011.
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Florestas (CNPF) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
URL |
Voltar
|
|
Registro Completo
Biblioteca(s): |
Embrapa Arroz e Feijão. |
Data corrente: |
20/05/2002 |
Data da última atualização: |
09/05/2012 |
Tipo da produção científica: |
Comunicado Técnico/Recomendações Técnicas |
Autoria: |
PRABHU, A. S.; GUIMARÃES, C. M.; BERNI, R. F. |
Afiliação: |
ANNE SITARAMA PRABHU, CNPAF; CLEBER MORAIS GUIMARAES, CNPAF; RODRIGO FASCIN BERNI. |
Título: |
Influência da época de plantio no controle da brusone em folhas de arroz de terras altas. |
Ano de publicação: |
2001 |
Fonte/Imprenta: |
Santo Antônio de Goiás: Embrapa Arroz e Feijão, 2001. |
Páginas: |
2 p. |
Série: |
(Embrapa Arroz e Feijão. Pesquisa em foco, 56). |
Idioma: |
Português |
Conteúdo: |
O presente trabalho objetivou estudar a influencia da época de plantio, da cultivar e do tratamento de sementes sobre a severidade da brusone nas folhas das novas cultivares de terras altas, melhoradas para alta qualidade de grãos. |
Palavras-Chave: |
Arroz de terras altas; Control; Controle; Controle de doença; Disease; Epoca; Oriza sativa; Terra Alta; Terras Altas; Terras altas: Brusone; Upland rice. |
Thesagro: |
Arroz; Brusone; Doença; Doença de Planta; Época de Plantio; Fungo; Glycine Max; Oryza Sativa; Plantio; Pyricularia Grisea; Rotação de Cultura; Soja; Variedade. |
Thesaurus NAL: |
rice. |
Categoria do assunto: |
-- H Saúde e Patologia |
URL: |
https://ainfo.cnptia.embrapa.br/digital/bitstream/item/59066/1/Foco-56.pdf
|
Marc: |
LEADER 01432nam a2200457 a 4500 001 1209835 005 2012-05-09 008 2001 bl uuuu u0uu1 u #d 100 1 $aPRABHU, A. S. 245 $aInfluência da época de plantio no controle da brusone em folhas de arroz de terras altas. 260 $aSanto Antônio de Goiás: Embrapa Arroz e Feijão$c2001 300 $a2 p. 490 $a(Embrapa Arroz e Feijão. Pesquisa em foco, 56). 520 $aO presente trabalho objetivou estudar a influencia da época de plantio, da cultivar e do tratamento de sementes sobre a severidade da brusone nas folhas das novas cultivares de terras altas, melhoradas para alta qualidade de grãos. 650 $arice 650 $aArroz 650 $aBrusone 650 $aDoença 650 $aDoença de Planta 650 $aÉpoca de Plantio 650 $aFungo 650 $aGlycine Max 650 $aOryza Sativa 650 $aPlantio 650 $aPyricularia Grisea 650 $aRotação de Cultura 650 $aSoja 650 $aVariedade 653 $aArroz de terras altas 653 $aControl 653 $aControle 653 $aControle de doença 653 $aDisease 653 $aEpoca 653 $aOriza sativa 653 $aTerra Alta 653 $aTerras Altas 653 $aTerras altas: Brusone 653 $aUpland rice 700 1 $aGUIMARÃES, C. M. 700 1 $aBERNI, R. F.
Download
Esconder MarcMostrar Marc Completo |
Registro original: |
Embrapa Arroz e Feijão (CNPAF) |
|
Biblioteca |
ID |
Origem |
Tipo/Formato |
Classificação |
Cutter |
Registro |
Volume |
Status |
Fechar
|
Expressão de busca inválida. Verifique!!! |
|
|