The loadPrismaData function triggers a feature selection and
data combination methods which subsequently remove duplicate entries for
efficient representation of the data. The
getDuplicateData rebuilds the data matrix with
explicit representation of all duplicate entries.