proposed system enhancements#56
Open
blakemertz wants to merge 4 commits into
Open
Conversation
Hardcoded CudaDeviceIndex='0' prevented targeting a specific GPU on multi-GPU workstations and made concurrent simulation runs impossible. Adds --gpu_id CLI flag (default 0) and passes it through to run_simulation() so each process can bind to an independent device. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Standard PDB format overflows at >99,999 atoms and >9,999 residues, causing MDAnalysis to lose all residue identity and corrupt atom selections. Switches save_pdb() to app.PDBxFile and updates the simulation loader to detect .cif vs .pdb by extension for backwards compatibility with existing topology files. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
Five bug fixes: cosolvent_names string coercion to list (prevented IndexError on single-name strings); generate_pymol_session list check before os.path.isfile (prevented TypeError); auto-discovery of .dx files when density_files=None; cmd.spectrum defaults to 'all' on empty selection (prevented PyMOL Empty expression error); VMD script trajectory type corrected from netcdf to dcd. Also extends traj_file and statistics_file to accept a list, enabling combined analysis of replicate runs via MDAnalysis ChainReader and pd.concat without changes to the rest of the pipeline. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
_build_box previously computed a single circumradius (max distance from center to any atom) and used it as the uniform box side, forcing a cube large enough to contain the protein's circumscribed sphere. For elongated proteins this wastes substantial solvent volume on the short axes, increasing atom count by 20-40%. Now computes an independent width per axis from the bounding-box extent plus padding: widthX/Y/Z = (maxRange - minRange) + 2*padding. The periodic box vectors, lowerBound, and upperBound are updated accordingly. Downstream geometry code (is_in_box, delete_edges_points, addSolvent) already operates per-axis and requires no changes. The no-receptor box_size path retains cubic geometry since a scalar radius is the natural input there. Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
blakemertz
marked this pull request as ready for review
April 9, 2026 20:08
Author
|
I may have identified a bug in the changes I made to analysis.py to handle multiple trajectories. My changes assumed an identical atom count across multiple systems and this is almost never the case. Sorting this out today, will get back to you with my updates. |
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Apologies for the delay in pushing these commits and drafting the pull request -- got hung up on a few other things. Proposing the following changes to the codebase: