Grubman et al 2019

[lcolladotor]

• Number of sets: 6 cell types * concordance (2 options) = 12 sets + direction 6 sets = 18 sets

• Note: snRNA-seq

• Direction available: Grubman.LogFC (AD vs Control a priori). There's not that many genes.

• Statistics available: No. But we have Concordance.

• [ ] For each cell type, use all genes (ignore Concordance)

• [ ] For each cell type, filter to Concordance == TRUE

• [ ] For each cell type, use all genes (ignore Concordance), and separate by direction.

Some might be too small also.

[shkwon17]

When Sang Ho manually checked the gene set of "Grubman et al", it seems that only one gene PTPRG overlapping with/enriched in the enriched gene set of the next_Ab category (blue box above in the heat map).

@sparthib could you double check the number of genes overlapping between the Grubman's microglia-related genes and the enriched genes of the next_Ab category?

Questions to be discussed with others:

Question 1) if it is really the single gene that only overlaps with the enriched gene set of the next_Ab category, can it drive that much statistically significant enrichment?

Question 2) if it is really the single gene, and the statistics are saying it is what it is, can I still make an interpretation that microglia or microglia-associated DEG(s) are enriched in the next_Ab category? or is it risky to make such an interpretation/statement?

[shkwon17]

Yes, it was the single gene, but we agreed we could make the interpretation/implication about microglia and their potential significance given this result!