Minor tweaks to sup fig 7

[gwaybio]

Main change is focus on AreaShape and Zernike features more closely in panels C and D, and also to combine old panels C and D into a single panel C

Supplementary Figure 7

Supplementary Figure 7. Comparing JUMP and Mitocheck nuclei feature spaces.

(A) Kolmogorov-Smirnov (KS) test results comparing JUMP and Mitocheck per common CellProfiler feature colored by specific CellProfiler feature group. The boxplot whiskers represent the interquartile range of 1,000 permutations of randomly subsampled JUMP single-cells from a single plate (JUMP Pilot plate BR00116991) compared to Mitocheck. Mitocheck and JUMP sample size is the same (n = 2,916). We show both raw and z-score normalized comparisons. (B) The same KS test results focused on AreaShape measurements, which showed the lowest differences in feature distributions across datasets. (C) Comparing variance of JUMP and Mitocheck for CellProfiler features. The dotted lines are the function y=x (anything below is a feature with higher variance in Mitocheck). The top plot showing all features, highlighting AreaShape, while the bottom plot focuses on 30 Zernike polynomial features.

[review-notebook-app]

Check out this pull request on 

See visual diffs & provide feedback on Jupyter Notebooks.

---

Powered by ReviewNB

[jenna-tomkinson]

Nice short PR! I added one clarifying question but overall, the minor changes LGTM! 🎉

[jenna-tomkinson]

I really like the change to focus panel C on AreaShape. I have just one comment:

1. I am confused by the normalized results in Panel C. Do we expect or want the variance to be higher for these features in Mitocheck, or what does this mean? I feel like this is most dramatic in the Zernike features, where the majority of the features are under the line meaning higher variance for Mitocheck.

Overall, this figure does convey the story well so great job!

[gwaybio]

Thanks!

Do we expect or want the variance to be higher for these features in Mitocheck, or what does this mean?

This panel shows that the variance across datasets is severely misaligned. We want variance to be similar, and it is most similar in the AreaShape and Zernike features.

I feel like this is most dramatic in the Zernike features, where the majority of the features are under the line meaning higher variance for Mitocheck.

Each point represents variance of a particular feature measured in 1,000 randomly sampled JUMP single cells compared to variance of all of MitoCheck. In other words, all CP features have 1,000 points in each facet of panel C.

Thanks for asking this question - I will work to make this more clear in the figure legend!