Integration of multiple single-cell multiomics samples for GRN construction #194
Replies: 1 comment 2 replies
-
Hi @Ajeet1699 Thank you for your kind words. Indeed, for the scRNA-seq part build a single h5 file containing all the samples. For ATAC, you can also generate a single cisTopic object for all the samples. See: https://pycistopic.readthedocs.io/en/latest/Cortex_pycisTopic.html for more information. For the GRN inference part. I would run it on all samples at the same time (not separately on disease and control samples). The reason for this is that the inference part needs contrasts to be able to find relevant networks. After running it once on all samples you can start comparing diseased and control states. I hope this helps. Best, Seppe |
Beta Was this translation helpful? Give feedback.
-
Hi,
Thanks for building the wonderful tool.
I have multiple single cell multiomics (same cells) samples (6 diseased and 6 control individuals) and I want to construct single GRN for each cell type. I am following
"https://scenicplus.readthedocs.io/en/latest/pbmc_multiome_tutorial.html"
which I understand was run only on one sample. What is the best way to integrate the multiple samples (cell ranger aggr
or bySeurat
) for building GRN? I can build an integrated .h5 object for scRNA seq data but for ATAC, there are separate "atac_fragments.tsv" files for each samples, how to handle these?And I am also wondering that will I have to build the separate GRNs for diseased and control samples? or is these a way to compare these in Scenic+?
Thanks in advance
Beta Was this translation helpful? Give feedback.
All reactions