Scholarlink

<p>Early-generation breeding trials (EGBTs) are usually performed as a part of selection process in order to achieve genetic improvement in plants. EGBTs examine many breeding lines with limited resources available in one or more environments. Replication of every line in every environment is often not possible due to resource constraints. Consequently, breeders opt for un-replicated trials, leading to less statistically reliable results. Partially replicated (p-rep) designs in multiple environments, where a proportion of the lines are replicated in each environment, offer an alternative. This paper presents two new methods for constructing series of p-rep designs that are suitable for multi-environmental trials. These designs are efficient and cost-effective in terms of resources as they require lesser replications and hence, they possess high application potential in EGBTs conducted over multiple environments. Further, an R-package “pRepDesigns” is developed that generates p-rep designs to facilitate the breeders for selecting appropriate designs. It is always possible to derive optimal PBIB designs from the proposed p-rep designs. A lower bound of mean square error for prediction of inbred performance is derived considering a random effects model under block set up.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Efficient Partially Replicated Designs for Multi-Environment Early-Generation Breeding Trials

  • 期刊论文

  • 发表日期：

    2025年11月24日

查看全文
  • L. N. Vinaykumar,
  • Cini Varghese,
  • Mohd Harun,
  • Sayantani Karmakar,
  • Eldho Varghese,
  • Vinayaka,
  • Seema Jaggi

摘要

Early-generation breeding trials (EGBTs) are usually performed as a part of selection process in order to achieve genetic improvement in plants. EGBTs examine many breeding lines with limited resources available in one or more environments. Replication of every line in every environment is often not possible due to resource constraints. Consequently, breeders opt for un-replicated trials, leading to less statistically reliable results. Partially replicated (p-rep) designs in multiple environments, where a proportion of the lines are replicated in each environment, offer an alternative. This paper presents two new methods for constructing series of p-rep designs that are suitable for multi-environmental trials. These designs are efficient and cost-effective in terms of resources as they require lesser replications and hence, they possess high application potential in EGBTs conducted over multiple environments. Further, an R-package “pRepDesigns” is developed that generates p-rep designs to facilitate the breeders for selecting appropriate designs. It is always possible to derive optimal PBIB designs from the proposed p-rep designs. A lower bound of mean square error for prediction of inbred performance is derived considering a random effects model under block set up.