Re: [Jmol-users] H bonds algorithm

2006-03-09 Thread Bob Hanson
Thanks, this was a very satisfying find. Why the difference: Jmol code is running through "polymers" not "chains", and I think the breaks define distinct polymers. Another difference is the minimum hydrogen bonding distance, which is 0.7 Angstroms in Jmol but 0.5 Angstroms in RasMol. I can't ima

Re: [Jmol-users] H bonds algorithm

2006-03-09 Thread Angel Herraez
Great work, Bob! I am not clear as to what happens with inter-chain bonds. Are they not being calculated? And in my example, they weren't even inter- chain, the chain was interrupted but was the same "A" chain. Rasmol gives Hbonds irrespective of the chain cut. -

Re: [Jmol-users] H bonds algorithm

2006-03-08 Thread Bob Hanson
Found the problem! The hydrogen bond calculation was incorrectly measuring the position of the peptide amino hydrogen. This is fixed and testable at http://www.stolaf.edu/people/hansonr/jmol/test/json/measure.htm prototype jar files are at http://www.stolaf.edu/people/hansonr/jmol/test/json

Re: [Jmol-users] H bonds algorithm

2006-03-08 Thread Bob Hanson
I've checked in a fix for this. It involved adding a check for maximum distance; seems odd that this was not there before. Miguel: comment? Hydrogens are calculated based on geometry, but only within chains. Prototype test is at http://www.stolaf.edu/people/hansonr/jmol/test/json/measure.htm

Re: [Jmol-users] H bonds algorithm

2006-03-08 Thread Bob Hanson
OK, I see them. Yes, this is wrong. I'll look into it. Hydrogen bonds are only about 1.8 - 2.0 Angstroms, but because the H atom itself is not shown, the distance between the two heteroatoms should be about 2.6-3.1 Angstroms. Someone can probably summarize how hydrogen bonds are determined in