search

schrodinger / pymol-open-source

Open-source foundation of the user-sponsored PyMOL molecular visualization system.
https://pymol.org/
Other
1.21k stars 284 forks source link

Bad Surface #300

Open wlincong opened 1 year ago

wlincong commented 1 year ago

Hi, Near Residue R181 (colored in orange)

https://github.com/schrodinger/pymol-open-source/assets/8821384/286574d7-92f6-40c3-8efd-59ced6470542

the SES surface for protein 7BK1 (pdbid) computed by Pymol has defects. Please see the attached video.

Is it a bug in pymol or happens to pymol rather frequently?

https://github.com/schrodinger/pymol-open-source/assets/8821384/83267c38-87b5-440d-8d1b-5c44d2bb6f8c

JarrettSJohnson commented 1 year ago

Hi, I'm not seeing what you are. Can you provide the list of steps how you got here?

wlincong commented 1 year ago

Dear Mr. Johnson, Thanks for the quick reply. I first added protons via "REDUCE -build ..." (from Duke) (the pdb I used is attached here as "7B1KFH.txt" file since gitHub refuses .pdb file) , and then computed the molecular surface via pymol with a choice of maximum quality. The defects appear near residue R181.

Thanks. Lincong 7BK1FH.txt

wlincong commented 1 year ago

Hi, Lincong again, I attached the SES for the same region by my program for your reference, BTW gitHub refuses .tif file, so I convert my original file to .jpg. Lincong 7BK1FH_SES_nearR181

JarrettSJohnson commented 1 year ago

Hi, yes. I can see the same issue from the first movie with surface_quality set to 1. Not sure if 2, or 3 look a bit better for you.

wlincong commented 1 year ago

Again thank for the reply.

FYI: I also checked with MSMS (via pmv) for the same region. MSMS gets away with possible defects by refusing to compute the top-open hole there. It seems the the pymol did try to compute/triangulate the hole but could not handle the complexity there.

Lincong

wlincong commented 1 year ago

Hi again, Last time I mentioned that "MSMS gets away with possible defects by refusing to compute ....". As the attached figure (ses_1zjcFH.pdb) shows though PyMOL in general could not compute/display correctly (i.e. in detail) a complicated SES region at least it tries. In the contrary, MSMS in general just ignores complicated SES features or tries to ignore the complicated SES features by increasing the radii of certain atoms. Please compare the 3 regions pointed by 3 arrows colored in cyan. The residue shown in color in the middle is 128. The MSMS figure is generated with a probe radius of 1.3, and a low_density of 50, a high_density of 100.

ses_1zjc.pdf the protein is 1zjc with protons added (1zjcFH.txt). 1zjcFH.txt