Hi Lily,
Now that the PR was merged I would like to provide more details about my proposal and ask you a couple of questions.
- Proposal.
I have been working on a python workflow to calculate membrane curvature out of trajectory + coordinate files from CG MD simulations. The calculation is performed using phospholipid head groups. The backbone of the code is there, it works, it's fast (it runs 2.5K frames in <1min with each frame comprising ~5K lipid headgroups) and it is automated to some extent. Essentially, it calculates the mean ($K_M$) and Gaussian curvature ($K_G$), and it plots the resulting values of $K_M$ and $K_G$ in contours or imshow using matplotlib. However, the next step I was considering was to translate the Python code from functional to object-oriented. And this sense, GSoC with MDAnalysis is a great opportunity to make this happen. Another point is that after working on the PR that was recently merged I have a better understanding of how pytest works, so I could also write the test for it.
In summary, I would like to write a curvature_analysis.py and test_curvature_analysis.py to calculate membrane curvature in terms of mean and gaussian curvature descriptors.
I think it would be super cool. Happy to share your feedback and opinions.
- Questions
As part of the applications process I should:
1) Introduce my self on the mailing list [Checked]
2) Close an issue of MDAnalysis. [Checked, issue #3209 and PR #3219]
4) Proposal Submission.
For the last one, I am not sure where to find the guidelines and I would appreciate your help with it in case you are ok with my proposal.
I'm open to discussing my idea in more detail and provide a timeline if you consider it appropriate.
Thanks!