Setting the standard for statistical analysis in plant breeding
In modern plant breeding today, breeders have an increasing amount of data to analyse and need to make rapid decisions.
Breeding View (BV) is a graphical user interface to a statistical analysis package that permits you to conduct phenotypic and genotypic analyses of the field trial observations that you upload to the Breeding Management System (BMS). It provides summary statistics and adjusted means of all traits and mixed model comparisons of genotypes and locations through a set of simple, visual and easy-to-use analytical pipelines and quality insurance measures:
"Breeding View enables easy and comprehensive data analysis and presentation of data very soon after completion of the study. It facilitates the work of breeders who will focus on plant selection rather than data management and analysis of technical aspects."
— Dr Ibnou Dieng, Head of the Data Integration and Biometrics Unit,
Africa Rice Center
These pipelines are presented as graphical workflows where you can choose to run the complete pipeline at once, or to run each separate element of the pipeline. Each element of each pipeline has a set of default statistical parameters which can be modified by the user before an analysis.
Single-site analysis – phenotypic analysis
Multi-site (genotype by environment) analysis – phenotypic analysis
Single trait (single environment) QTL analysis using external sources of data.
Results of each analysis are saved in a time stamped folder in your personal workspace or, for the phenotypic analysis, can be saved to the BMS database for later review or further analysis. The phenotypic analyses are fully integrated into the BMS so that data can be selected in the BMS and sent to Breeding View for analysis, and resulting means and summary statistics saved back to the BMS database, allowing database searches based on germplasm performance within and between locations.?
A quality assurance step in the single-site analysis displays a list of potential outliers for each trait by environment combination and allows the user to re-run the analysis with specified outliers treated as missing values to verify the effect of these aberrant values on the conclusions of the study.