You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
{{ message }}
Repository navigation
Feature request: Processing on multiple sets of FASTQ files #13
I frequently work with single-cell multi-omics data (mostly scRNA-seq + CITE-seq, but scATAC-seq is also expected), and it is known that different modalities for the same samples requires separate sequencing libraries for each modality, which means there will be multiple sets of FASTQ files per sample. What I'm looking for in splitcode is the ability to reassign barcodes for several sets of FASTQ files uniformly such that multi-omics data can seamlessly integrate. Is it possible to do this?
Are you talking about the final barcodes that are created when you use --assign? You'd like those barcodes to carry over from, say, your scRNA-seq data to your CITE-seq data?
Or, do you mean, supplying scRNA-seq data and your CITE-seq data in a single splitcode command?
Activity
Yenaled commentedon Mar 22, 2024
Hmm, not sure I understand.
Are you talking about the final barcodes that are created when you use --assign? You'd like those barcodes to carry over from, say, your scRNA-seq data to your CITE-seq data?
Or, do you mean, supplying scRNA-seq data and your CITE-seq data in a single splitcode command?
Perhaps a simple example would help.