Episode 5, Exercise 1 - Histogram Normalization Resources

[wangliyuan72]

Histogram normalization

I find the following resources helpful on this topic:
CellProfiler team made blog post in 2022 and sc-forum post on how to normalize high-content cell painting dataset. They prefer RobustMAD (median) over standardized (mean) method. From their post, the code for the function here. What is also nice is that they included Plate-Layout Effects

The third one is Percentiles based normalization from resource posted by DigitalSreeni organoid analysis ipynb and found scikitimage's example.

Learning objective(s)

Covers the Labelling with Connected Components.

Maybe we opt in for Quantiles or percentiles method?

Volunteer(s)

@wangliyuan72 @marcodallavecchia @terezbelinova Let me know if I should add more info to the different sections in this post. Thanks.

[marcodallavecchia]

Hey @wangliyuan72, thanks a lot for the ideas and resources, I find them quite useful!

I would agree that they are relevant for the histogram normalization learning objective, however, I would say this issue currently describes more an additional resource and material we can draw inspiration from rather than an exercise proposal for this specific objective.

This is very much encouraged, so keep them coming! However for house-keeping reasons, I will update some of the labels and Issue description to be fit the content.

We can definitely open another Issue with a more specific exercise proposal that uses this external material and link this issue to the proposal itself. You are more than welcome to give it a try!