Modularity and integration analysis of rice seed using MINT ver. 1.0b software

The study of phenotypic integration concerns the modular nature of organismal phenotypes. The concept provides for a rationale on the observation that certain subsets of phenotypic traits show particularly high levels of association over development and/or evolution. The rice seed consists of morphological parts that can be argued to be a good model for modularity and integration analysis thus this study was conducted. Patterns of phenotypic integration and modularity in the mature seeds of weedy rice using the Modularity and Integration Analysis Tool (MINT) ver. 1.0b were used in this investigation. A total of 8 variational modularity models in the shape data were tested using the γ* (Gamma*) test for goodness of fit (GoF) statistics. This was used to compare the observed and expected covariance matrices and implemented based on a total of 1,000 replicates. Jackknife support values for each model were also computed using γ* as the GoF statistic. Results revealed that the best fitted model showing the mature rice seed is organized into two statistically independent modules, one controlling for the lemma and the other for the palea. This model has a standard gamma value of 0.18575 and a P value of 1 based on a Monte Carlo test with 1000 replicates. The observed partitioning of the co-variation of morphological data into statistically independent blocks or units suggests the presence of modules in the rice seed.