What are some good uses of [sorted].bam files after RNA-seq?

Question

[O.rka]

I have my first single celled dataset and there are about 2000 single cell transcriptomes. I usually keep the sorted.bam files [STAR -> samtools sort] but I'm a little limited on space at the moment.

I'm wondering if I should keep the BAM files after I use featurecounts for mapping? As a backup, I've put all of the read identifiers into a list.

What good use cases can I use for RNA-seq bam files after they have been counted?

In metagenomics, I use these for coverage estimates and binning pipelines but I haven't thought of a good reason to keep these other than having files for backup in case I lose the counts table.

[Ram]

We use sorted BAMs to generate read depth data (wig files). Not sure if it's good for anything else.

Answer 1

[ATpoint]

For single-cell data you normally don’t need it for anything once you have the counts unless you plan to try something like single-cell variant callign or similar strategies. You can try and compress the files to CRAM, that should save you some space. Or just buy a big external harddrive. These are almost trivially cheap these days.

[Santosh Anand]

Ditto. Storage space is cheap, but computation (time) is costly. So if you could buy an external HD, move the bam files there as you could never guess with certainity if they could be needed in future or not

[Ram]

External HDs, especially if they're a platter, stand a high risk of corruption. I'd recommend a NAS server loaded with vibration protected high performance disks or even Amazon Glacier if the user can afford their cold storage.