How to concatenate andata objects when you have citeseq data

General
1

Hello,

I have .h5 files that contain gene expression and cite seq data. I am trying to confirm that the way I imported them seems correct:

Sample1 = sc.read_10x_h5(‘E:\Sample1_filtered_feature_bc_matrix.h5’, gex_only=False)
Sample1 .var_names_make_unique()

Sample2 = sc.read_10x_h5(‘E:\Sample2_filtered_feature_bc_matrix.h5’, gex_only=False)
Sample2 .var_names_make_unique()

Sample3 = sc.read_10x_h5(‘E:\Sample3_filtered_feature_bc_matrix.h5’, gex_only=False)
Sample3 .var_names_make_unique()

Sample1.obs[‘sample’]="Sample1 "
Sample2.obs[‘sample’]="Sample2 "
Sample3.obs[‘sample’]="Sample3 "

merge into one object.

adata_merged = Sample1 .concatenate(Sample2, Sample3, join=‘outer’)
adata_merged.obs_names_make_unique()

The adata_merged object seems ok and I can see the cite seq in adata_merged.var but I just want to make sure that it was done properly.

Thank you!

2

Yes looks correct. Calling var_names_make_unique seperately can have unexpected side effects, if you have data with different var_names.

3

Thank you for replying. What side effects should i look into? The .var from the gene expression is the same but i have different hashtags in some of the samples so there are a few different names.

4

If the order is different of genes in different datasets, you can end up with different names for the same gene and same name for different genes (it attaches -1 etc to the gene names).

5

Thank you for replying. Is there a better way to concatenate then?
I have not noticed -1 in the gene names. I do notice that it creates a gene ids-0, gene ids-1 etc in adata.var.
I see that on the barcodes it adds a -0, -1 etc at the end of the barcodes. Would that create an issue?