Skip to main content

Table 6 Summary of ProDiGe variants

From: ProDiGe: Prioritization Of Disease Genes with multitask machine learning from positive and unlabeled examples

Name Disease kernel Sharing of disease gene information across diseases
Prodige1 K Dirac No sharing.
Prodige2 1 + K Dirac Uniform sharing.
Prodige3 K phenotype Sharing weighted by phenotypic similarity.
Prodige4 K Dirac + K phenotype Sharing weighted by phenotypic similarity and disease identity.
  1. We propose four variants, which differ in the way they share information across diseases, as summarized in the third column of the table. The second column shows the kernel for diseases used by each variant to achieve the sharing of information.