Fig 6. Classification model for distinguish between early RA and normal based on RPKMs.

The model consists only of the single rule LXN > 3.8 AND CXCL8 > 0.04 -> early RA. It corresponds to an accuracy of 92% at the 10-fold cross-validation (p-value 2.02*10⁻¹³). Variables for model-generation were pre-selected upon the intersection of single-variable comparisons. This pre-selection weakens the cross-validation as it is no part of it. This model is only intended for a distinction between early RA and normal as simple as possible based on gene expression. In total there are 84 samples. The RPKM values are cut at 10 and at 1, respectively.