next up previous
Next: Reliability scoring Up: Results and Discussion Previous: Prediction of solvent accessibility

Blind test of prediction methods

The results of a blind test of the new Jnet and other prediction methods that apply multiple sequence alignments for prediction are shown in Table 5, and Figure 4. Predictions were made for 406 proteins not used to develop the methods (see Methods). On this set of 406 proteins, Jnet gave an average accuracy of 76.4%. This is 3.1% better than the best previous method (PHD, 73.3%) and 1.8% better than the Jpred [34] consensus method.

Table 7 summarises a closer investigation of the differences between PHD and the Jnet method. From these figures it is clear that while the Jnet method is more accurate than PHD, the $\beta$-strand state is not predicted any more accurately by Jnet than PHD (0.1%). Most of the improvement is coming from the clearer delineation of the helix and coil states, (1.6 and 4.1% respectively).



James Cuff
2001-06-29