search

liulab-dfci / TRUST4

TCR and BCR assembly from RNA-seq data
MIT License
274 stars 47 forks source link

Hashed samples in 10X #150

Open learning-MD opened 2 years ago

learning-MD commented 2 years ago

Hi, first of all, thank you for this fantastic resource!

I have 5' scRNA-seq data on PBMCs from 40 patients that have been hashed with antibodies - 5 samples tagged and run in a lane. Are there any specific adjustments that need to be made before running TRUST4 on the large BAM file (or FASTQ files) of these hashed samples?

Thanks in advance!

mourisl commented 2 years ago

I don't think there will be a problem with that. The memory usage might be large due to the run contains immune repertoires from 40 samples.

Just curious, is there a mapping to know which barcode corresponds to a sample or patient? Thanks.

learning-MD commented 2 years ago

Thanks! Yes, there are feature barcodes to identify what corresponds to what sample. I think I'll just have to run one hashed sample to see how it works out and take it from there.