Starting from 10X matrix file

[kokp2000]

Hi,

I have problem with integrating datasets using 10x output files.
It gives me the error message " Raw data must have both row (cell) and column (gene) names" although my data has valid gene names (var_names) and cell ID (obs_names).

Do you know how to solve this problem?

[Rbass3]

when create_liger reads each adata object, it expects var_names and obs_names (within each adata object) to have a "name" attribute which indicates that var_names represents gene names and that obs_names represents barcodes. so, you need to set the "name" attribute for var_names and obs_names to be "gene_names" and "barcodes" respectively. running this code on each adata object in the list fixed the issue for me:

Hope this fixes the issue