CL - Help File

Last Update August 19, 1997

General

Conformational Likeness (CL) is a methodology for finding 3-D homology between proteins of known structure. It is suggested you review the tutorial for an idea of the capabilities of CL. If you need further information refer to the citations.

Complete Database Search

Each of the options to be selected is discussed below:

Entity

Polypeptide chain designated 4HHB:A - the structure represented by the PDB ID 4HHB and the chain identifier A. If the polypeptide chain is not identified use an underscore such as 5EBX:_

Entity Search/Fragment Search

Either a search is perfomed using the whole entity as the probe or a fragment of the entity. Fragments are specified by starting and ending residues numbers. Numbers are given relative to the N terminus which is 1. CAUTION this numbering scheme may be different to that adopted by the authors and found in the PDB file.

Property Subset

This selects a group of properties against which to performa similarity search. A search against all properties is very time consuming and not necessarily useful (see below)

The name of the group is broken in two parts: the first part is the type of alignment; the second part is the search criteria used. For example LocalDist implies: Local == based on the pentapeptide value; Dist == local distances between CA's and between CA's and the center of mass. First parts of the names:

• "Local" - based upon local pentapeptide values.
• "Seq" - based upon the PAM matrix as defined by Pearson.
• "Feat" - miscellaneous collection of features.
• "Neigh" - based on pentamer values with respect to surrounding decamers. before and after in the sequence i.e. sequence neighbors.
• "Envir" - based on residues withing a 25 Å radius.
• "Topol" - a combination of "Neigh" and "Envir".
• "Prot" - based on whole protein.

Second parts of these names:

• "Dist" - set of bond lengths, other CA lengths and distances to centers of mass.
• "Angle" - set of bond angles between CA atoms and centers of mass.
• "Twist" - set of dihedral angles between CA atoms and centers of mass.
• "Surface" - exposure and polarity of pentamers as defined by Lee and Richards
• "PhiPsi" - main chain dihedral angles
• "SecStr" - secondary structure calculated by the method of Kabsch and Sander using the alphabet {H,G,I,T,B,E,S}
• "PAMatrix" - mutation matrix as defined by Pearson.
• "Shape" - volume, number of chemical bonds, molecular weight.
• "Freq" - amino acid frequencies in a 25 Å radius of each pentamer.
• "ChouFasman" - classic alpha and beta coefficients.
• "Thang" - plane angles between CA's and CA's and center of mass of protein and pentamer.
• "Ring" - dihedral angles between CA's and protein center of mass.
• "Round" - variety of dihedral angles.

Length Criterion

One of five options can be used to speed the search by restricting the difference in protein length between the query probe and any target proteins. Only proteins with a length difference less than the given percentage value from the query protein will be reported.

Likeness Criterion

A range of 0.0 to 1.0 specifies the degree of conformational likeness. Only matches with likeness greater than the threshold will be reported. The larger the value the less similarity matches will be reported.

Detailed 1-on-1

Results

Conformational likeness is shown as a 2-D plot with degree of likeness highlighted according to the spectrum at the bottom left of the plot. The first entity is shown on the x axis and teh second on the y axis.

Likeness (red) indicates an alignment, using dynamic programming, for the overall likeness based upon the features selected.

Align (purple) indicates an alignment, using dynamic programming, for the best alignment of the sequences based on the features selected. Thus, in most cases where some topological likeness is found, the red line will expand the purple.

At center is the sequence alignment based on conformational likeness values selected thus this is an alignment of sequences resulting from the alignment of structures NOT the reverse.

At bottom are the superimposed entities color coded by degree of superposition. This is a stereo pair and can be viewed with stereo glasses. Alternatively, the superimposed coordinates can be downloaded or output as a Rasmol script.