def createAnnData from lpy.py

[Chris-lang478]

Dear authors,

Thank you for your help again. I have a question about your code and data. In lpy.py, I found this function:

def createAnnData(folderlist, prefix, souporcell_folderlist = None, souporcell_genodico = None, autoinclude=["percent_mito", "log2p1_count", "n_genes"], min_cell_per_gene_allowed=3, min_gene_per_cell_allowed=500, sample_obskey = "sample_names",doqcplots=False, doinspect=False, mitogeneprefix="MT-", do_log2_normalize=True)

There is a line that says:

if do_log2_normalize: adata.raw = adata.copy() adata.X = np.log1p(adata.X) / math.log(2)

However, I found that adata.raw is empty from the processed matrices (h5ad file) from [www.reproductivecellatlas.org] . So I think the processed matrices have no original counts.

My question is: Is my understanding correct? If so, how can I get the original counts from the RCA website? Or do I have to download and process the raw data files myself? I appreciate your time and attention.

[luzgaral]

Hi,

The objects here https://www.reproductivecellatlas.org/non-pregnant-uterus.html
should contain original raw counts as adata.raw.

Hope this helps.

Best
Luz

[Chris-lang478]

@luzgaral
Thanks for your reply. And I just downloaded the objects from the website as you mentioned. However ,this is the results of adata.raw. I am a newer user for scanpy, and I am counfused about this result. Doesn’t it mean empyty?