Issue with 5' technologies with single-ends
TomKellyGenetics opened this issue · comments
As mentioned by @Davidwei7 (#12) there are known issues with STRT-Seq. An ex-colleague from RIKEN has also contacted us about supporting this technique. I've suggested using the (undocumented) "SC5P-R1" chemistry to run Cell Ranger. However, the latest version of UniverSC has automated checks that require both R1 and R2 files to exist, assuming that all technologies are paired-end.
As a work-around, I suggest copying R1 to and R2 FASTQ file, hard-trimming the barcodes from R2 and running 3' paired-end chemistry in Cell Ranger with a custom barcode whitelist of STRT-seq presets.
The stable solution is to refactor the input checks for UniverSC to conditionally allow single-end FASTQ inputs (missing R2 files) for 5' scRNA-Seq data and the associate technologies (i.e., STRT-seq, SCRB-seq, ICELL8). I won't have time to do this for a while but I'll note here publicly that this is a known issue and is being addressed.
Technology "strt-seq" has been updated to default to single-end 5' chemistry "SC5P-R1" and allows --read1 to give single-end FASTQ input without a ---read2 file input.
I've tested these parameters on data from:
GSE29087 (SRR212477) published by Islam et al. (2011): https://genome.cshlp.org/content/21/7/1160.full
In principle manual input of SC5P-R1 chemistry for any technology should allow the same parameters.