There are no minimally invasive diagnostic metrics for acute kidney transplant rejection (AR), especially in the setting of the common confounding diagnosis, acute dysfunction with no rejection (ADNR). Thus, though kidney transplant biopsies remain the gold standard, they are invasive, have substantial risks, sampling error issues and significant costs and are not suitable for serial monitoring. Global gene expression profiles of 148 peripheral blood samples from transplant patients with excellent function and normal histology (TX; n = 46), AR (n = 63) and ADNR (n = 39), from two independent cohorts were analyzed with DNA microarrays. We applied a new normalization tool, frozen robust multi-array analysis, particularly suitable for clinical diagnostics, multiple prediction tools to discover, refine and validate robust molecular classifiers and we tested a novel one-by-one analysis strategy to model the real clinical application of this test. Multiple three-way classifier tools identified 200 highest value probesets with sensitivity, specificity, positive predictive value, negative predictive value and area under the curve for the validation cohort ranging from 82% to 100%, 76% to 95%, 76% to 95%, 79% to 100%, 84% to 100% and 0.817 to 0.968, respectively. We conclude that peripheral blood gene expression profiling can be used as a minimally invasive tool to accurately reveal TX, AR and ADNR in the setting of acute kidney transplant dysfunction.