engleski

Main Effect Meta Principal Component Analysis (ME-MetaPCA) as the Tool of Choice for Processing Typical Horticulural Metadata

In numerous horticultural researches the authors’ interest is related to changes in fruit characteristics caused by treatment in one or more genotypes. However, the results are often contradictory. The general significance and the size of the treatment effect can be estimated reliably by meta-analysis using metadata. In horticultural research there are no known developed procedures for the use of meta-analysis which represents a challenge to generate it in order to process the typical multilayer data profile originating from multiple sources and multiple measurements scales. The answer for this typical horticultural metadata setting is application of Main Effect Meta Principal Component Analysis (ME-MetaPCA). ME-MetaPCA represents specific combination of Meta-analysis and Principal Component Analysis adjusted specifically for the use on horticultural data. The goal of this paper is to demonstrate the application and usefulness of this novel technique through simulations and analysis of a typical horticultural data - in apple as a model crop. Data were modeled in terms of plant growth regulator treatment effect researched in several studies on multiple apple varieties. The treatment effect is measured in multiple fruit characteristics on multiple measurement scales. Simulation model is realistically composed of five different studies involving five apple varieties and measuring six fruit characteristics providing total of 5×5×6, i.e. 150 study level data, which with recorded means, standard deviations and sample size for both treatment and control formed a matrix of 900 data. Besides the standard forest plot as a result of the meta-analysis, the results of the ME-MetaPCA are displayed in the form of biplot in principal components space. This provides straightforward and simple in-one-glance overview of the definite general effects of the studied treatment on i. individual fruit characteristics, ii. studied apple varieties and iii. grouping patterns with grouping rules and exceptions.

biometrics, meta-analysis, horticulture

nije evidentirano