Hello Pascal, > Great - I've been using the Marcou's C++ program. Works fine. > I don't use it so often, but I'll look at your version and try it when > I have some time - I'm not a good programmer, but it's, as you said, > much easier for me with python...
Wow, great makes sure to check the current svn too (http://code.google.com/p/pyplif/source/checkout), and let me know if you would like to develop this project with me :) > I personnaly use Surflex or Flexx outpout (actually only site.mol2 + > poses.mol2). I would really appreciate it if you can send me some of the output from those tools so my PyPLIF can support wider range of output, my current PyPLIF is designed to work automatically for PLANTS's output. My next plan is to support the output from autodock vina. > Were you able to reproduce the results from the Marcou's paper? It's still on the benchmarking process, hopefully we can publish it in July or August :) > What do you think of an additionnal score for comparing new ligands to > the reference? Let's say the reference has 2 interactions, and that > you have a new ligand with exactly the same 2 interactions and a third > interaction: this ligand would have a lower tanimoto score, but it > would be interesting to know that all reference interactions are still > reproduced. I'm sorry I'm not sure I get what you're saying. Are you saying that the program should be able to measure the tanimoto coefficient of more than 1 ligand at the same time? If so, it's already done actually. As you can see in my project page, there are Promazine and Amitypriline as the ligand test (and others too but not shown). > And what about giving different weights to different types of > interaction? E.g. 2 reference interactions: one H-bond and one > hydrophobic. One ligand with only the H bond. One other ligand with > only the hydrophobic contact. Same tanimoto coefficient for the 2 > ligands. But in my opinion the ligand with the polar interaction would > be more interesting. Any thought about that? Yup, I do thought about that, though I'm a bit confused on how to do that, perhaps the bit for weighted interaction should be doubled, so for example when there's hydrogen bond instead of using just 1 or 0 we can use 11 or 00. Another thing that we can do is giving weight to certain residue, I think that will be interesting :) -- View this message in context: http://forums.openbabel.org/PyPLIF-Another-Open-Babel-related-project-tp4530931p4532034.html Sent from the General discussion mailing list archive at Nabble.com. ------------------------------------------------------------------------------ Better than sec? Nothing is better than sec when it comes to monitoring Big Data applications. Try Boundary one-second resolution app monitoring today. Free. http://p.sf.net/sfu/Boundary-dev2dev _______________________________________________ OpenBabel-discuss mailing list OpenBabel-discuss@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/openbabel-discuss
