Kappa Value and Agreement Rate algorithms

  • 1. For each gene/feature, convert their are values to binary
    • new value(proteinomic data) = 1
      if old value is greater than 1,
      else new value = 0
    • new value(Microarray data) = 1
      if old value is not a null,
      else new value = null
  • 2. N = total number of genes/features
  • 3. YY = values of both samples are 1s
    NN = values of both samples are 0s/missing
    YN = value of sample1 is 1, and value of sample2 is 0/missing
    NY = value of sample1 is 0, and value of sample2 is 0/missing
  • 4. Pct_YY = YY / N
    Pct_NN = NN / N
    Pct_YN = YN / N
    Pct_NY = NY / N
  • 5. Agreement rate of sample1 and sampe2 = Pct_YY + Pct_NN
  • 6. Kappa Value of Sample1 and sample2
    Observed agreement = Pct_YY + Pct_NN
    Chance Agreement = (Pct_NN + Pct_NY)*(Pct_NN + Pct_YN) + (Pct_NY + Pct_YY)*(Pct_YN + Pct_YY)
    Kappa Value = (Observed agreement - Chance Agreement)/(1 - Chance Agreement )

Edit | Attach | Print version | History: r5 < r4 < r3 < r2 | Backlinks | View wiki text | Edit WikiText | More topic actions...
Topic revision: r4 - 02 May 2005, JoanZhang
 

This site is powered by FoswikiCopyright © 2013-2022 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Vanderbilt Biostatistics Wiki? Send feedback