Andreas wrote: > browsed up the code to the point to where Helices are calculated and > HBond partner - offsets are stored in mainchainHbondOffsets[]. > > Offsets are required to determine Helices. For sheets it is needed to > store all HBonds in another array. Additionally, each amino > acid can be donor or acceptor of two H-bonds. DSSP calls this > "bifurcated" H-bonds. Basically energyMin1, indexMin1 and > energyMin2,indexMin2 in bondAminoHydrogen should be stored and are > needed for method "bond" in the pseudocode below.
I have checked in code which mimics this. For each donor residue the energy values and indexes of the acceptor are saved. > pseudocode to calc sheet H-Bond pattern: > for a in aminos: > for b in aminos: > // testing for parallel : > if ( bond(a+1,b) && bond(b,a-1)) || > ( bond(b+1,a) && bond(a,b-1)) > found parralel bridge > > // testing for antiparallel : > if ( bond(a,b) && bond(b,a)) || > ( bond(a+1,b-1) && bond(b+1,a-1) > found antiparallel bridge. > > and the code for bond: > > boolean bond (i,j) > if ( ( i.acceptor1.index = j && i.acceptor1.energy < cutoff )|| > ( i.acceptor2.index = j && i.acceptor2.energy < cutoff )) > return true ; > else return false, Once I find these bridges, it is not clear to me which residues should be tagged. At first I thought that only the donors should be tagged as 'participating in a sheet'. But then I decided that we needed to tag the recipients as well. > according to the code in AminoPolymer "cutoff" could be -500 which is > also the cutoff used during the calculation of the energy. OK > then we have got the basic elements of sheets: H-Bond bridges. > Several bridges are connected to a sheet if they involve consecutive > residues. In using 1CRN.pdb none of the bridges are consecutive (unless I tag both the donors and recipients). Even then, there are things that look like to me that they should be tagged, because their two neighboring residues are participating in the sheet. But, if I allow gaps in the tags then non-sheet things get included. Please advise Miguel ----- Open Source Molecular Visualization www.jmol.org [EMAIL PROTECTED] ----- ------------------------------------------------------- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_idU88&alloc_id065&op=click _______________________________________________ Jmol-developers mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/jmol-developers
