Prediction of the subcellular location
Up to 20 individual prediction programs contribute to the prediction of the subcellular location (see Resource > Subcellular location for prediction program details). Notice that not all programs perform predictions for all four locations considered (chloroplast, mitochondrion, secretory pathways, nucleus).
Consensus prediction of the subcellular location (AramLocCon)
The individual predictions for each protein are combined to a built-in consensus prediction using the Bayes' theorem. The resulting consensus score is based on the ratio of the posterior probabilities of both hypotheses, i.e. that a protein is targeted and that it is not targeted given the positive prediction results. The corresponding likelihoods of the hypotheses were estimated using a set of training plant proteins with known subcellular locations. The consensus prediction method has been published in:
Schwacke R, Fischer K, Ketelsen B, Krupinska K, Krause K
Comparative survey of plastid and mitochondrial targeting properties of transcription factors in Arabidopsis and rice
Mol Genet Genomics. 2007 Jun, 277(6): 631-46. (PubMed)