FORECAST Web Server

Introduction

FORECAST is an automated protein fold recognition server combining profile-profile alignment and support vector machine. An input sequence is PSI-BLASTed against NCBI NR database(<90%) to make its profile. The profile is aligned with templates to calculate feature vectors and the corresponding feature vectors are classified by SVM method.

Features

  • Sensitive and specific fold identification
  • Provide accurate alignment maximizing MaxSub score for further processes such as structure modeling

Template Library

5955 protein domains from SCOP 1.67 (<40%)
Discontinuous domains and Membrane proteins are excluded.

References

  • Sangjo Han, Byung-chul Lee, Seung Taek Yu, Chan-seok Jeong, Soyoung Lee, Dongsup Kim. (2005) Fold recognition by combining profile-profile alignment and support vector machine, Bioinformatics, 21, 2667-2673.

Program Distribution

If you want to run this program on your computers, contact us by forecast@pbil.kaist.ac.kr