Research data from field trials and data analysis, comparing recurrent haploid selection, mass selection, full-sib selection, S2 selection and doubled haploid line selection in a sweet corn population

This dataset contains the research data underlying the dissertation “Comparison of selection methods in a sweet corn population”. The dissertation investigated the effects of different selection methods on marketable yield, plant development traits, ear quality traits and related agronomic traits in an open-pollinated sweet corn population.

The data were generated in field trials and selection experiments conducted over several years. The dataset includes raw data, cleaned datasets, R scripts, statistical outputs, graphical data and additional photographic documentation. The selection methods covered include recurrent haploid selection, positive mass selection, full-sib selection, S2 line selection, selection of doubled haploid lines produced via colchicine-induced chromosome doubling, and selection of spontaneously doubled haploid lines.

For Chapter 2, the dataset contains data from five cycles of recurrent haploid selection and five cycles of positive mass selection. The initial population and the selected cycles were evaluated in multiple environments under organic and conventional management conditions. Recorded traits include plant development traits such as plant number and vigour, ear quality traits such as ear length, ear diameter, tip fill, ear shape, colour and row number, as well as yield traits such as total yield, marketable yield, total number of ears and number of marketable ears.

For Chapters 3 and 4, the dataset contains data from the evaluation of one cycle of full-sib selection, one cycle of S2 line selection, and selection based on doubled haploid and spontaneously doubled haploid lines. The data include selection-year data, evaluation-year data, cleaned datasets from different trial locations, adjusted datasets, means used for graphical outputs, and R scripts used for data preparation, data cleaning and statistical analysis.

The analyses were carried out mainly in R. Statistical methods included analysis of variance, estimation of variance components, Tukey tests, calculation of heritability, adjustment of data for experimental design or environmental effects where applicable, and identification of influential outliers using Cook’s distance. The R scripts included in the archive document the main steps of data preparation and analysis and indicate which data files were used for the respective analyses.

The dataset is intended to make the results of the dissertation transparent, reproducible and verifiable. It provides the basis for the tables, figures and conclusions presented in the dissertation and associated publications. No personal or sensitive data are included.