Remove duplicate reads in single cell bam file

Question

[emre.cto]

Is there a tool to remove duplicated reads from single cell bam file? I am currently dealing with single cell atac-seq file and apparently people have used their own custom scripts for that. E.g this article:

https://www.sciencedirect.com/science/article/pii/S0092867418308559?via

[GenoMax]

Did your BAM file come from 10x data? Do you have multiple cells in a single BAM file? Are you referring to deduplicating the reads based on their UMI or plain sequence?

[emre.cto]

Yes it is 10x data and I have multiple cells in a single BAM file. I want to remove the duplicates based on plain sequence, I do not have UMI sequence.

[GenoMax]

Barcodes and UMI's should be in the BAM file. They are encoded by the BC and CR tags (for RNAseq data).

Since this is 10x ATACseq data can you post a couple of lines of your BAM?

You can use @Pierre's solution here to split the file by cell barcodes (BC tags) : A: Splitting a bam file by unique optional TAG field

You may need to covert the data back to fastq if you want to use a solution like clumpify to remove duplicates based on sequence. If by dedupe you mean based on existing alignment position then please confirm that.