Hi Shengfu,
Glad to hear that Chimera does a good job of hydrogen addition relative to other packages.  It can still get things wrong with extremely complex ring systems or poor starting coordinates, but seems pretty reliable otherwise.  I’m glad you like it!
One problem here is that Python memory consumption may not go down when in the middle of script execution.  You can in fact run Chimera from a terminal with no graphical interface by giving the --nogui option, as described here: https://www.cgl.ucsf.edu/chimera/docs/UsersGuide/options.html .  Therefore you could break your computation into batches of 500 or 1000 molecules and possibly work around the memory problem that way.
There is no standalone binary for adding hydrogens within Chimera.  There is a Python module (AddH), but it relies heavily on Chimera’s built-in atom typing, and which is therefore difficult to run outside the context of Chimera.

—Eric

Eric Pettersen
UCSF Computer Graphics Lab


On Nov 20, 2019, at 8:56 AM, Williams, Joanne <Joanne.Williams@ucsf.edu> wrote:



JoAnne Williams | Executive Analyst to:

 

Matt Jacobson, Professor and Department Chair
Pam England, Professor
Zev Gartner, Professor
Tom Ferrin, Professor
Department of Pharmaceutical Chemistry

 

Wallace Marshall, Professor
Center for Cellular Construction | Department of Biochemistry & Biophysics

 

600 16th Street, Room N372 | UCSF MC2200 | San Francisco, CA 94158 
Phone: (415) 418-3079 | Email: joanne.williams@ucsf.edu

From: 周盛福 <zhousf@szbl.ac.cn>
Sent: Tuesday, November 19, 2019 1:33 AM
To: chimera <chimera@cgl.ucsf.edu>
Subject: To add hydrogen atoms for a dataset
 
Hi all,

    I have used UCSF Chimera for several years and find that Chimera has a perfect behavior in adding hydrogen atoms for small molecules. It is easier to add wrong hydrogen atoms onto nitrogen heterocycles groups with other software, including openbabel, MGLTools, PyMol, etc. Now I want to add hydrogen atoms for a dataset, which contains more than 10 thousands ligands for virtual screening. And a great number of compounds in this dataset have nitrogen heterocycle groups.
    So I want to add hydrogen atoms for all these compounds by Chimera.
    When I try to run a script in Favorites > Command line by read a MOL, addh and delete the MOL in cycle. I found that the memory usage of Chimera keep growing although I have delete the previous MOLs.

    My question is:
        How can I avoid this problem while I want to handle numerous molecules with Chimera?
        If there is a way to run Chimera in a real terminal without run Chimera in graphic interface?  Or
        Which bin file in Chimera_HOME can add hydrogen atoms for a molecule?

Sincerely look forward to your reply.
Thanks

Best
Shengfu Zhou
2019.11.19
_______________________________________________
Chimera-users mailing list: Chimera-users@cgl.ucsf.edu
Manage subscription: http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-users