Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

[WIP] Feature/sliding window denoising #7

Draft
wants to merge 24 commits into
base: main
Choose a base branch
from

Conversation

jspaezp
Copy link
Collaborator

@jspaezp jspaezp commented Jun 24, 2024

This PR actually bundles several things ...

  1. Refactors a lot of the aggregation/denoising/dbscan logic.
  2. Removes unused compiler flags
  3. Changes the order in which frames are aggregated (split+aggregated in groups by isolation window before denoising, not during)

@jspaezp
Copy link
Collaborator Author

jspaezp commented Jul 8, 2024

TODO:

  • Add unit testing for frame slices (make sure offsets + conversions are handled correctly)
  • Implement denoising with window:
    • Implement IntenseAtIndex (which also entails implementing a proxy 'point' struct), AsNDPointsAtIndex and NDPointConverter for slice frames.
    • Implement a SliceFrameRTWindow struct and -> AsNDPointsAtIndex + NDPointConverter for that.
    • Implement aggregator for that 'point' struct ...
    • Replace element wise denoising ->
      denoise_frame_slice(
      &x,
      &self.ims_converter,
      &self.mz_converter,
      &self.dia_frame_info,
      self.min_n,
      self.min_intensity,
      self.mz_scaling,
      self.max_mz_extension,
      self.ims_scaling,
      self.max_ims_extension,
      with window-wise denoising.
  • Chores + maintainability:
    • Rename dbscan functions.
      • _dbscan -> dbscan_label_clusters?
      • dbscan_generic -> IndexAndDbscanAggregate ? ... maybe split it as a Runner with .convert + .index + .label + .aggregate + .refine
  • Revise the generic annotations + param names for dbscan to more sensible 'names'
  • Make explicit what distances are in projected dimensions and which are not.
  • Consider bundling the extra filter fun with the DistantAtIndex trait.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

1 participant