search

JSB-UCLA / scDesign2

An interpretable simulator that generates realistic single-cell gene expression count data with gene correlations recapitulated
MIT License
23 stars 4 forks source link

Autochoose: Is there a way to tell which gene got what marginal distribution? #6

Open emucaki opened 1 year ago

emucaki commented 1 year ago

Hello,

I'm just going through the tutorial, which uses the "Autochoose" function when selecting a marginal distribution. I was wondering if there is any way to find out from the output which individual genes got which distribution? I see the output has "marginal_param1" and "marginal_param2", but I don't think there's a way to figure it out from there.

I was interested to see if the frequency of marginal distributions chosen would inform us about our dataset (i.e. if a certain distribution was being selected more often than others).

Thank you!

EDIT - so I found in a manual the following text regarding the 3 columns in 'marginal_param': "the ZI proportion, the dispersion parameter (for Poisson, it's Inf), and the mean parameter.}"

So if I'm reading this right: 1st column: 0, 2nd Column: 'Inf' = Poisson 1st column: 0, 2nd Column: Not 'Inf' = NB 1st column: Not 0, 2nd Column: 'Inf' = ZIP 1st column: Not 0, 2nd Column: Not 'Inf' = ZINB

Is that accurate? What is the difference between "marginal_param1" and "marginal_param2"?