Skip to main content

Table 1 The input data used by the ANNs in the different ProfNet methods.

From: Improved alignment quality by combining evolutionary information, predicted secondary structure and self-organizing maps

Method

plain profiles

Prob_score score

pred. sec. str.

SOM locations

ProfNet

Y

N

N

N

ProfNet_SS

N

Y

Y

N

ProfNet_SOM

N

Y

N

Y

ProfNet_SS_SOM

N

Y

Y

Y

  1. In ProfNet the "plain profile", i.e. the amino acid frequencies are used as the only input. In ProfNet_SS the sequence similarity is instead measured using the Prob_score measure and in addition six values from PSI-PRED predicted secondary structure probabilities are added. ProfNet_SOM is identical but the PSI-PRED values are replaced with the output from the SOM classification and in ProfNet_SS_SOM both outputs are combined.