Wrong exposure selected in the Manually Curate Exposures scatter plot

Exposure Curation
1

1
11487×924 305 KB

2
21490×925 303 KB

While selecting exposures manually, I noticed that the micrograph selected is sometimes not internally consistent. For example, in the first image, when I hover over the point shown in the scatter plot, it tells me that this is exposure 1409 (see image 1). Upon rejecting, another seemingly random point in the graph turns red (not shown in my images), but the point I’m hovering on remains blue. Surprisingly, this now changes it’s identity to another micrograph! Now it says that the original point was actually 1408 (see image 2). This can be repeated many many times, and seems to always walk through the index by -1. It seems to be pulling the right info for the micrograph as the images change, but it’s the plot that is incorrectly associating the micrograph number with the coordinates shown. This is further evidenced by the fact that the coordinates don’t match the CTF fit resolution or ice thickness in the hover display. Has anyone else seen this behavior before? I’m using cryosparc version 4.3.0. Thanks in advance.

2

Hi @Justin ,

Welcome to the forum and thanks for your post!

From my interpretation, it looks like you’re hovering over points and able to see the details for that particular exposure. It’s only until you click a point does it get marked as the selected exposure (marked in blue on the table) and becomes available to toggle rejection status via keyboard shortcut or button.

Can you confirm that when you click and select an exposure, you can correctly toggle rejection status?

- Suhail

3

Hi Suahil, thanks for your reply.

Yes, that’s correct. When I click on the point, I can correctly toggle rejection status. But the plot was giving me the incorrect micrograph details, so I was toggling the rejection status of some other micrograph. So, if I zoomed out in the plot you could see some other point turning red.

-Justin

4

Hi @Justin ,

Thanks for the clarification, we’re taking a closer look at this now.

- Suhail

5

Hi @Justin,

Thank you for the detailed report and screenshots.

We have identified the cause of this bug and will have it fixed in our next release.

- Kelly

1 Like
6

Good to hear! I’m glad it was helpful. Thanks!

  • Justin