Skip to main content
. 2006 Apr 18;4(1):48–55. doi: 10.1016/S1672-0229(06)60016-8

Table 1.

Comparison of the Localization Performance of PredSL and TargetP Tested by Five-fold Cross-validation and Self-consistency*

Predictor set Overall accuracy (%) Category Sensitivity Specificity MCC
A. PredSL (cross-validation/self-consistency)
Plant 86.7/88.3 cTP 0.90/0.90 0.80/0.91 0.82/0.88
mTP 0.89/0.96 0.87/0.81 0.84/0.85
SP 0.96/0.95 0.92/0.89 0.91/0.90
other 0.70/0.72 0.86/0.95 0.74/0.79
Non-plant 87.1/92.5
mTP 0.88/0.91 0.84/0.96 0.80/90.5
SP 0.94/0.95 0.91/0.91 0.89/0.90
other 0.80/0.92 0.86/0.91 0.77/0.88

B. TargetP (cross-validation/self-consistency)
Plant 85.3/90.4 cTP 0.85/0.96 0.69/0.78 0.72/0.84
mTP 0.82/0.88 0.90/0.95 0.77/0.88
SP 0.91/0.94 0.95/0.94 0.90/0.92
other 0.85/0.85 0.78/0.87 0.77/0.84
Non-plant 90.0/92.2 mTP 0.89/0.92 0.67/0.72 0.73/0.79
SP 0.96/0.97 0.92/0.95 0.92/0.95
other 0.88/0.90 0.97/0.97 0.82/0.86
*

The PredSL datasets for plant proteins consist of 249 chloroplast sequences, 250 mitochondrial sequences, and 253 secreted proteins’ sequences, whereas for non-plant proteins the datasets consist of 366 mitochondrial sequences and 370 secreted proteins’ sequences. The TargetP datasets for plant proteins consist of 141 chloroplast sequences, 368 mitochondrial sequences, and 269 secreted proteins’ sequences, whereas for non-plant proteins the datasets consist of 371 mitochondrial sequences and 715 secreted proteins’ sequences.