Comparison of digital image analysis using elliptic Fourier descriptors and major dimensions to phenotype seed shape in hexaploid wheat (Triticum aestivum L.)

Digital image analysis (DIA) is widely used for describing plant organ shape. However, the various types of shape descriptors that can be generated using DIA may identify different loci in genetic analyses. The purpose of this study was to evaluate two different DIA approaches to quantifying wheat seed shape for exploring trait correlations and quantitative trait loci (QTL) mapping. Phenotypic data were produced using the software programs ImageJ (National Institutes of Health, USA, http://rsbweb.nih.gov/ij/) and SHAPE (Hiroyoshi Iwata, http://lbm.ab.a.u-tokyo.ac.jp/similar or equal to iwata/shape/). ImageJ generates measures of length, width, perimeter, and area that can be used to describe dimensions of objects, whereas SHAPE generates elliptic Fourier descriptors (EFDs) to capture shape variation such as roughness, asymmetric skewing, or other two-dimensional aspects not encompassed by axes or distinctions in overall object area. There were significant differences in the results of the QTL analysis depending on the DIA software used. The use of EFDs to characterize horizontal measures of seed shape in wheat identified more QTL with higher LOD scores than length to width ratio. Additionally, the entire three dimensional shape of the seed described using two images in different orientations was shown to identify seed shape QTL that co-located with flour yield (FLYLD) and would go undetected based solely on a two dimensional image of the seed. Both methods identified QTL for length, width, thickness, and vertical perimeter that were co-localized with QTL for FLYLD.