This is a test version of Biostars. For the public version, visit https://www.biostars.org.
Retrieving individual cells from scRNA-seq bigWig files?

Hi,

I have a bigWig file containing single-cell RNA-seq data of ~50k individual cells (public dataset obtained from GEO). I would like to run some tests analyzing the profiles of individual cells. However, I don't seem to be able to retrieve individual cells from the bigWig file but, instead, I obtain a bedGraph/coverage track of all cells combined (I used UCSCtools bigWigToWig + bedops wig2bed).

Is it really possible to obtain individual cells from the bigWig file or I will only get a pool of all samples?

Thanks,

Txema

single cell rna-seq scrna-seq bigwig
• 2,573 views • link
updated by GouthamAtla GouthamAtla • written by txema.heredia • txema.heredia

2 answers

No, this is not possible. You will have to download the raw data or (if available) a count matrix to analyze things yourself.

• 0 views • link
written by ATpoint ATpoint

Oh, dang!

It seems that the raw BAMs are not available, but I could find what seems to be the count matrix.

This file has 3 columns. From what I understand, the first column is the gene_ID, the second column is the cell_ID, and the third column is the ¿count? ¿count of reads overlapping the gene? ¿partially of fully covering the gene?

I am interested in analyzing partial/complete or blocked transcription. Can I work with this count matrix or I would need the raw data to do it?

Thanks,

Txema

• 0 views • link
written by txema.heredia • txema.heredia

I would personally always start from raw data unless there is processed data and a crystal-clear explanation how these data have been processed computationally. With raw data I mean fastq files.

• 0 views • link
written by ATpoint ATpoint

Its impossible to get individual cells from bigwig. You need to have original bam file (if its 10x) or individual bam files (if its well-based).

• 0 views • link
written by GouthamAtla GouthamAtla

Log in to answer this question.

More posts like this