Skip to main content
. 2016 Oct 4;11(10):e0163711. doi: 10.1371/journal.pone.0163711

Table 3. Comparison of classifications obtained using cohort based normalization against Exlab and InLab reference based normalization.

The classifications are compared in terms of accuracy, Cohen’s weighted κ, and Pearson’s correlation coefficient r all supplied with 95% CIs. The comparisons in the first and last three columns are based on the ExLab and InLab reference based normalization method, respectively. For ABC/GCB classification, results from InLab or Exlab classification with the elasitic net classifier is compared against ABC/GCB classes for cohort normalized data obtained using both Wrights Bayes classifier and the elastic net classifier.

ExLab RMA pre-processing InLab RMA pre-processing
Accuracy Cohen’s κ Pearson’s r Accuracy Cohen’s κ Pearson’s r
ABC/GCB (Wright)
 CHEPRETRO .89 (.80, .94) .89 (.79, .98) - .97 (.88, 1.) .97 (.90, 1.) -
 MDFCI .63 (.52, .73) .52 (.40, .64) - .72 (.59, .83) .71 (.55, .86) -
 IDRC .67 (.63, .71) .62 (.56, .67) - .84 (.80, .87) .82 (.77, .86) -
 LLMPP R-CHOP .83 (.77, .87) .82 (.74, .89) - .88 (.83, .92) .88 (.82, .93) -
ABC/GCB
 CHEPRETRO .88 (.79, .94) .87 (.78, .97) .999 (.998, .999) .98 (.91, 1.) .98 (.93, 1.) 1. (.999, 1.)
 MDFCI .69 (.59, .78) .68 (.53, .82) .998 (.998, .999) .98 (.91, 1.) .98 (.85, 1.) 1. (.999, 1.)
 IDRC .65 (.61, .69) .62 (.57, .68) .986 (.983, .988) .93 (.91, .95) .93 (.90, .96) .993 (.991, .994)
 LLMPP R-CHOP .82 (.77, .87) .82 (.74, .89) .999 (.999, .999) .94 (.90, .97) .94 (.90, .98) .991 (.988, .993)
BAGS
 CHEPRETRO .58 (.47, .69) .56 (.28, .84) - .78 (.65, .88) .74 (.33, 1.) -
 MDFCI .54 (.43, .64) .48 (.17, .79) - .80 (.68, .89) .83 (.30, 1.) -
 IDRC .52 (.47, .56) .41 (.32, .50) - .79 (.75, .83) .79 (.62, .96) -
 LLMPP R-CHOP .56 (.49, .62) .53 (.36, .70) - .88 (.82, .92) .88 (.60, 1.) -
REGS
 CHEPRETRO .73 (.68, .78) .71 (.64, .77) .934 (.920, .946) .84 (.79, .88) .83 (.76, .89) .992 (.990, .994)
 MDFCI .60 (.55, .65) .55 (.48, .61) .824 (.788, .855) .90 (.86, .94) .89 (.83, .96) .997 (.996, .997)
 IDRC .52 (.49, .54) .33 (.30, .36) .660 (.635, .685) .85 (.84, .87) .84 (.81, .86) .981 (.979, .983)
 LLMPP R-CHOP .58 (.54, .61) .50 (.46, .54) .810 (.786, .831) .90 (.87, .92) .89 (.85, .92) .992 (.990, .993)