search

liulab-dfci / TRUST4

TCR and BCR assembly from RNA-seq data
MIT License
278 stars 49 forks source link

Spatial Transcriptomics Question #105

Open leeanapeters opened 2 years ago

leeanapeters commented 2 years ago

Hi, thank you for the great package! We have used this pipeline with scRNAseq data with great success.

I am wondering if it would be possible to apply TRUST4 to 10x visium data? Do we need to make any adjustments to the pipeline? In particular, it seems like we would need to incorporate the spatial barcode for this to work and be able to associate with the corresponding gene expression data.

Any input you have would be much appreciated!

Thanks!

Leeana

mourisl commented 2 years ago

I never tested the 10x visium data before. Theoretically, TRUST4 should work without any adjustment. One thing I can think of is that since there are multiple cells in one gel (same spatial barcode) in 10X spatial platform, and TRUST4 only picks one representative pair of chains for each barcode, directly using the results may not be ideal. TRUST44 reports those "secondary" chains in the barcode_report file, and you may consider those results too. Furthermore, in the barcode_report file, TRUST4 will pick one cell type (B, abT, gdT) for each barcode, and ignore the CDR3s from other chains. So, you may need to use the raw _cdr3.out file to comprehensively analyze the spatial data.

Furthermore, if 10x visium data is 3' RNA-seq, then TRUST4 would have very low sensitivity.