After the internal validation by leave-n-out approaches suggested that the full classifier (all 12 compounds) works with good predictivity, we proceeded to independent validation approaches based on external data sets.
A novel 14 probe-set signature was developed to classify the tumors internally with 93% accuracy, and this was successfully validated on an external data-set with 94% accuracy.
The number and value of external data sets will rise.
We report the derivation of a novel and useful gene predictor validated both on an internal and an independent external data-set, implying its generalizability.
We validated the use of the MGMT-STP27 model in an external data-set of 50 glioblastoma (E-GBM) analyzed on the HM-27K [ 7].
In the following section, the results comprising sophisticated stochastic ZWD models, the effect on station coordinates, and comparisons to external data sets are presented and analyzed.
We will test these hypotheses using data from GP practices linked to external data sets only.
Judging from the attained statistical results, our derived QNAR models have an acceptable overall accuracy and robustness, as well as good predictivity on the external data sets.
The obtained model was validated on two external data sets consisting of in vivo human jejunal permeability coefficients (Peff) and absorption rate constants (Ka).
The identified 51-gene signature was then evaluated on independent external data sets predicating distance metastasis in breast cancer patients with lymph-node-negative tumours [ 4].
