Running long-read and paired data

[cprovido]

Hi,
Is it possible to use SPLASH on MinION-generated fastq files? I have 10 fastqs corresponding to 10 different cell types that were sorted prior to sequencing. Was hoping to use SPLASH to identify transcripts unique to a cell type so I considered running SPLASH on each fastq separately and independently analyze results.

Second, I have another dataset of bulk RNA sequencing with all paired files. I read in #13 to just use the R1 files and I'm wondering why that is? What are advantages of using R1 only than the concatenated files?

[roozbehdn]

Hi, in principle you should be able to run SPLASH on long-read data as well, though we have not yet systematically evaluated its performance on such data.
Regarding your second questiuon, splash requires at least two input samples to perform differential analysis. My suggestion is to run SPLASH on all FASTQ files together and then look at the calls, which would essentially be celltype specific (as each fastq comes from a distinct celltype).