Skip to content

ID/DIV plotting changes#18

Merged
scharch merged 3 commits intoscharch:masterfrom
ShawHahnLab:for-15-plotting-changes
Oct 28, 2022
Merged

ID/DIV plotting changes#18
scharch merged 3 commits intoscharch:masterfrom
ShawHahnLab:for-15-plotting-changes

Conversation

@ressy
Copy link
Contributor

@ressy ressy commented Sep 17, 2022

This proposes two changes to the ID/DIV plotting:

  1. Expands the plot area to -1 on the x-axis and 101 on the y-axis, to ensure that points lying directly on an axis (no divergence from germline or 100% identity to a mAb reference) still get plotted.
  2. Adds an alternate plotting method to plot binned sequence counts per ID/DIV region instead of estimated sequence density.

That second one can be enabled by giving --plotmethod binned to the script, where --plotmethod original is still the default since the other method does result in somewhat different-looking plots. This should fix #15 by making sure the points in that example are always visible.

@scharch how's this look? Just let me know if any of this should be implemented differently.

Expands the x axis' lower bound to -1 and y axis' upper bound to 101 so
that points lying directly at 0% divergence or 100% mAb identity are
still plotted (see warning about removal of data outside limits in
?ggplot2::scale_continuous) and expands available selection area in
interactive plot to match
Adds a --plotmethod to the get_island script, and code to plot_all, to
support a method for binning points to plot instead of using KDE.
@scharch
Copy link
Owner

scharch commented Oct 28, 2022

@ressy Thanks! Have you tested how changing the boundaries of the plot affects the behavior of gglocator? There's always been some wobble where the point placed doesn't seem to hit exactly where you clicked...

@ressy
Copy link
Contributor Author

ressy commented Oct 28, 2022

Just judging by eye (and memory) I don't think changing the plot boundaries changed that behavior either way, though yeah, now that you mention it I think I've just subconsciously adjusted to that slight offset with the clicking... and I think that's still the same. (I had thought it might just be something odd about my remote X11 setup, but I guess not!)

@scharch scharch merged commit 72f0d4c into scharch:master Oct 28, 2022
@ressy ressy deleted the for-15-plotting-changes branch November 21, 2023 20:21
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.

ID/DIV plots don't shown sequences for certain edge cases

2 participants