Allele-Specific Expression (ASE) from scRNA-seq dataset

[singcell]

Hello,
Thank you for developing this nice tool.
I'm interested in analyzing allele-specific expression (ASE) in my single-cell RNA-sequencing (scRNA-seq) data. This data comes from offspring tissues generated by crossing two different mouse strains with known distinct single nucleotide variants (SNVs) and polymorphisms. What is the best approach for conducting this analysis using SCreadCounts tool? Do you have a pipeline designed for this purpose?

Thank you

[edwardsnj]

Hi @singcell . SCReadCounts can certainly give you cell-specific and allele-specific UMI (or read)-counts given prior knowledge of the distinguishing loci. You would prepare your SNV file to describe the loci and their reference and alternative alleles. You could then observe the number of UMI (or reads) with the reference and alternative allele with respect to each cell barcode. While crude, this measurement of expression doesn't need to be normalized for transcript length etc. since only those reads overlapping the loci of interest are counted. That said, its probably best to consider this a relative measurement between alleles specific to each locus. Does this help?

[singcell]

Hello edwardsnj, I appreciate your helpful reply. Is it possible to incorporate SCReadCounts tool to make this processing easier for me who does not have bioinformatics background. Since I am working with imprinted genes which are either expressed from only one allele and not other, therefore I am not worried about quantification. I just want to know maternal vs paternal expression. I will process my data and will post here if I have any questions.

Thank you.

[edwardsnj]

Do you know the loci on the reference genome that indicate maternal vs paternal expression? If so, you can make the required SNV input file. Lines look like this:

CHR# position

Note that you should align your reads to the reference genome using STARsolo or CellRanger first.

The output of the tool will provide you with the number of UMIs/Reads in each cell that contain each of the alleles at each provided locus.

@AneliaHorvath may have other suggestions too.

[singcell]

edwardsnj Thank you. Some of the imprinted genes are known and as you indicated I can update the input file. I also want to find novel imprinted genes by analyzing the allele specific expression of 3' transcripts generated from 10x genomics.
Thank you