Prediction of Protein Functional Domains from Sequences Using Artificial Neural Networks

(Downloading may take up to 30 seconds. If the slide opens in your browser, select File -> Save As to save it.)

Click on image to view larger version.

Figure 1.
Figure 1.

Variables used to characterize domain similarities. AverageBLAST similarity score (AVS) versus number of significant BLAST similarities (NSD) plot for Trypsin domains (red circles) from SBASE 7.0 (Murvai et al. 2000a). Each red circle represents a domain sequence. The AVS andNSD values are computed from the BLASTsimilarities between the sequence and other members of the group. Blue triangles represent domain sequences that are not Trypsin domains but still produce a significant BLAST similarity score with some of the Trypsin domains. The database versus database comparison was carried out with BLAST (Altschul et al. 1990), by use of a significance threshold of 0.8 and a minimum score of 32. (Inset) Schematic representation of the empirical distributions (see text for details). Values of P Formula (NSD), and P Formula (NSD), are simply read out from the precomputed empirical distributions. A similar procedure is followed for the AVS scores. The resulting six values are the input parameters of the ANNs.

This Article

  1. Genome Res. 11: 1410-1417

Preprint Server