SUPPLEMENTARY DATA

 

fRMSDPred: Predicting local rmsd between structural fragments using sequence information.

By Huzefa Rangwala (rangwala AT cs DOT umn DOT edu) & George Karypis (karypis AT cs DOT umn DOT edu)

Computer Science Department at the University of Minnesota-Twin Cities

DATASETS (FOR JOURNAL PAPER VERSION)
DATASET DSA (FOR RELIABILITY PREDICTION) is avaialble here.
DATASET DSA   (FOR fRMSD ESIMATION) is available here.
DATASET DSR (FOR RELIABILITY PREDICTION AND FRMSD ESTIMATION) is available here.
DATASETS (FOR CSB 2007 VERSION )
DATASET EASY  is available here.
DATASET HARD is available here.

                                                                    CODES


The programs we used in this study were

  1. support vector machines (SVM Light) for building base classifiers
Please feel free to contact us at (rangwala@cs.umn.edu) to get the codes for generating the
various feature vectors & reliability definitions.
Extra Tables
In extra tables
we have some results (not included in paper due to lack of space).
1. Regression performance evaluated using root mean square error and correlation coeffecient on residue-pairs selected
    from random positions rather than sequence based aligned.
2. Why the equal weighting on the individual kernels for the fusion kernels was a good choice.
Extended version of the paper with more results and charts will be available shortly.

Figure 1 in the paper is available here.
Figure 2 in the paper is available here.

Corresponding Author: Huzefa Rangwala
Last Updated on Mar 26, 2007