Skip to main content

Advertisement

Table 3 Multi-location prediction results on the No-PROSITE-GO, No-PROSITE, and No-GO versions of the dataset, averaged over 25 runs of 5-fold cross-validation, for the combined set of single- and multi-localized proteins, using our system

From: Protein (multi-)location prediction: using location inter-dependencies in a probabilistic framework

  Dataset F 1 F1-label Acc
SVMs (without using dependencies) No-PROSITE-GO 0.75 (± 0.04) 0.66 (± 0.02) 0.70 (± 0.04)
Our system (using dependencies) No-PROSITE-GO 0.78 (± 0.05) 0.72 (± 0.07) 0.73 (± 0.05)
SVMs (without using dependencies) No-PROSITE 0.77 (± 0.01) 0.66 (± 0.02) 0.72 (± 0.01)
Our system (using dependencies) No-PROSITE 0.80 (± 0.01) 0.75 (± 0.02) 0.75 (± 0.01)
SVMs (without using dependencies) No-GO 0.76 (± 0.03) 0.67 (± 0.03) 0.71 (± 0.03)
Our system (using dependencies) No-GO 0.79 (± 0.04) 0.72 (± 0.08) 0.74 (± 0.04)
  1. The table shows the F1 score, the F1-label score, and the overall accuracy (Acc) obtained from SVMs without using location inter-dependencies and from our system, which uses location inter-dependencies. Standard deviations are shown in parentheses.