wenweixiong / MARVEL

38 stars 9 forks source link

MARVEL for 30000+cells from ICELL8 platform #38

Closed caochch closed 3 months ago

caochch commented 3 months ago

Dear,

I used the ICELL8 scRNA-seq platform and sequenced ~30000 cells, with each cell having 10^6 reads. Which strategy do you suggest to quantify the splicing junction count for my scRNAseq data?
Of note, ICELL8 is a plated-based platform.

Looking forward to your reply Best wishes Changchang Cao

wenweixiong commented 3 months ago

Given that it's plate-based platform and high depth sequencing (10 ^ 6 = 1,000,000 reads per cell), I would recommend using the plate-based approach for splicing analysis.

caochch commented 3 months ago

That will be a huge project, running STAR for 30000 times to count splice junction reads

wenweixiong commented 3 months ago

You may run STARsolo on the master BAM file (consisting of all cells, similar to cellranger's possorted_genome_bam.bam) to obtain the splice junction x cell matrix, and then further format the matrix for MARVEL's plate-based analysis.

caochch commented 3 months ago

Thanks a lot for this help. Best wishes.