Skip to main content

Table 3 Performance comparison on 5618 Euk-mPLoc protein dataset

From: Gene ontology based transfer learning for protein subcellular localization

Subnuclear location size GO-TLM
   SP SE MCC
Acrosome 17 0.9286 0.7647 0.8422
Cell 53 0.8085 0.7170 0.7593
Centriole 64 0.8958 0.6719 0.7737
Chloroplast 501 0.9681 0.9681 0.9650
Cyanelle 85 1.0000 0.9882 0.9940
Cytoplasm 1060 0.6356 0.7983 0.6475
Cytoskeleton 74 0.2500 0.0877 0.1431
Endoplasmic 364 0.7453 0.7790 0.7461
Endosome 89 0.6591 0.3867 0.4999
Extracell 640 0.7895 0.6402 0.7034
Golgi 254 1.0000 0.9231 0.9607
Hydrogenosome 13 0.7872 0.5968 0.6825
Lysosome 80 1.0000 0.4615 0.6789
Melanosome 13 0.6000 0.3103 0.4295
Microsome 31 0.9349 0.8865 0.9020
Mitochondrion 535 0.8071 0.8145 0.7689
Nucleus 1333 0.9412 0.8696 0.9032
Peroxisome 97 0.8059 0.7781 0.7658
Plasma 725 0.9260 0.8410 0.8694
Spindle pole 36 0.8750 0.4118 0.5995
Synapse 15 1.0000 0.5385 0.7334
Vacuole 102 0.9176 0.8571 0.8851
Overall Accuracy 80.38%
Baseline: Euk-mPLoc model 67.40%
Baseline: Fuzzy K-NN model 62.25%