Before complex prediction, we evaluated the optimal values of DPClus parameters by comparing the predicted complex set with the experimentally determined set of 144 reference complexes.
For survival prediction, we evaluated samples from three independent cohorts of patients with primary breast cancer: the GEICAM/9906 and CALGB/9741 phase III clinical trials (Citron et al, 2003; Martín et al, 2008), and the METABRIC data set, which is a UK/Canadian cohort of nearly 2000 primary breast cancers with transcriptomic and outcome data (Curtis et al, 2012).
For the ancestral location prediction, we evaluate the results by computing the average geographical distance between predicted locations and true locations in simulations (prediction error).
To place MeSHOP comparisons in relationship to a top literature-based candidate gene prediction tool, we evaluated predictions for Alzheimer disease (AD -gene relAD -geneps by MeSHOP comparelationshipseading tool.
Although miRGate uses established and well-known prediction algorithms, we evaluated the predictions obtained by those methods against a dataset of experimentally validated targets.
We proposed a learning-based approach for LTE KPI performance prediction, and we evaluated it by using data obtained from realistic urban small cell network simulations.
In our prediction benchmarks, we evaluated the Clarke-based cancer profiles using exactly the same procedure we used to evaluate the ISOpure-estimated cancer profiles, as outlined below in the 'Gene signature identification and testing' section.
After training process with the whole training dataset, in order to select the best prediction model, we evaluated diagnostic abilities based on the Bayesian information criterion (BIC) in the internal validation dataset.
Using a series of well-described class prediction algorithms, we evaluated the capacity of our custom CodeSet to accurately classify samples within the training series into the correct subgroup.
Besides evaluating the performance of complex prediction, we also evaluate the performance of edge classification, in which the edge weights are used to classify edges as co-complex or non-co-complex edges.
