ShinyCore: An R/Shiny program for establishing core collection based on single nucleotide polymorphism data

Background: Managing and investigating all available genetic resources are challenging. As an alternative, breeders and researchers use core collection—a representative subset of the entire collection. A good core is characterized by high genetic diversity and low repetitiveness. Among the several available software, GenoCore uses a coverage criterion that does not require computationally expensive distance-based metrics.

Results: ShinyCore is a new method to select a core collection through two phases. The first phase uses the coverage criterion to quickly attain a fixed coverage, and the second phase uses a newly devised score (referred to as the rarity score) to further enhance diversity. It can attain a fixed coverage faster than a currently available algorithm devised for the coverage criterion, so it will benefit users who have big data. ShinyCore attains the minimum coverage specified by a user faster than GenoCore, and it then seeks to add entries with the rarest allele for each marker. Therefore, measures of genetic diversity and distance can be improved.

Conclusion: Although GenoCore is a fast algorithm, its implementation is difficult for those unfamiliar with R, ShinyCore can be easily implemented in Shiny with RStudio and an interactive web applet is available for those who are not familiar with programming languages.