Skip to main content

Advertisement

Table 3 Number of genes for training and success rates of LDA

From: A computational approach to candidate gene prioritization for X-linked mental retardation using annotation-based binary filtering and motif-based linear discriminatory analysis

Set Analyzed Parameter 10 kb (Genes) 50 kb (Genes) 100 kb (Genes)
Training and test set of genes in XAR τ 0.96 0.43 0.4
  Success in XLMR 100% (19) 100% (15) 100% (9)
  Success in non-XLMR 45% (84) 91% (40) 96% (26)
Training and test set of genes in XCR τ 0.87 0.75 0.62
  Success in XLMR 87% (38) 100% (16) 100% (7)
  Success in non-XLMR 52% (257) 82% (141) 96% (74)
  1. τ is a tuning parameter, which was selected to maximize the sum of correct classification rates for XLMR and non-XLMR sets