Comparing the distribution of median scores between L1000 and Lincs Cell painting Consensus datasets

[gwaybio]

I am pasting @AdeboyeML's analysis performed in #2 (comment) so that we can continue a targeted discussion in a clean github issue:

---

@shntnu @gwaygenomics

Comparing the distribution of median scores between L1000 and Lincs Cell painting Consensus datasets

- Major points

• 213 MOAs (Mechanism of actions) present in both Cell painting and L1000 Level-5 data are compared based on the distribution of their median scores.

• During alignment of MOAs in L1000 with the MOAS in Cell painting, I realized that MOAs found in the same broad sample in both L1000 & Cell painting data are partly named differently i.e. the naming of same MOAS in both are not consistent.

Results -- MODZ Consensus dataset

Scatter plot btw L1000 vs Lincs Cell Painting median scores per dose

• Median scores in cell painting data are more spread out and have more extreme median values than L1000 data.

Distribution of median scores in L1000 and Cell Painting Data per dose

Distribution on a dose-by-dose basis

- I am still trying to figure out the reason behind the relationship between the p-value and median scores (null distribution) in #2 (comment)

[gwaybio]

@AdeniyiML - this is beautiful.

I see a couple things from this analysis:

• Cell Painting gets more MOAs "right" and more MOAs "wrong" than L100
• Only a handful of MOAs have really strong groups in L1000
• Dose doesn't really impact things all that much

A couple things to keep in mind:

• we know that MOA annotations are wrong - thinking that they aren't will hurt us trying to interpret these results
• It makes sense to me that morphology will group things more tightly than gene expression. There are many ways to alter a gene expression profile that can manifest into similar morphology.
• I am most interested in following up on some groups of MOAs that are bad in one tech and good in the other (the off-diagonal)👇

[AdeboyeML]

@gwaygenomics

I am most interested in following up on some groups of MOAs that are bad in one tech and good in the other (the off-diagonal)

• MOAs with high median scores in cell painting and low scores in L1000: