Paired layout, but one fastq file

Question

[Andy]

Good morning everyone,

I found data from GEO, GSE115469, and the author stated that it is in a paired fastq layout. However, I have only found one fastq file. I am wondering how to handle this fastq file because I need to re-run the cell ranger process.

Thanks Andy

[Ram]

If you're dealing with bulk data (technically possible even in single cell data), Paired End reads can also be interleaved and stored in a single FASTQ file. Always examine file content - that's part of the sanity check.

[GenoMax]

That would be unusual, since this is single-cell data.

[Ram]

OP's post does not mention single-cell, which is why I suggested this possibility. For future users that might run into this problems, I think it's important to understand that number of files may not be the most reliable indicator of SE/PE nature of sequencing.

[GenoMax]

GEO accession listed in original post is for single cell dataset.

[Ram]

I initially didn't bother looking up the GEO entry. I assumed it was bulk so the option did not cross my mind. I'll move my answer to a comment. I wish OP would have mentioned it - if I assumed bulk, so might other people that actually run into this problem in the bulk context.

Answer 1

[Andy]

I understand why now, the author only shared bam file.

[Ram]

I've moved your comment to an answer and added an answer of my own. Please accept your answer and optionally mine too to mark your post as resolved.

[GenoMax]

You can recreate the fastq files using a tool provided by 10x genomics called bamtofastq (LINK). This will properly recreated the CB+UMI - R1 file and R2 RNA read file.

[Andy]

Yes, bamtofastq do solve the problem. And this gives me correct fastq files.

[Sky]

I am also having a similar issue where there is only one fastq file but no bam file available so from my understanding I can't use bamtofastq. Is there another solution? I have attached the link to the dataset I am trying to download: https://trace.ncbi.nlm.nih.gov/Traces/?view=run_browser&acc=SRR14667226&display=metadata.

I have tried the following: fasterq-dump SRR14667226 --include-technical -S
fasterq-dump SRR14667226
fastq-dump SRR14667226 --split-3 --skip-technical

[GenoMax]

There is a BAM file available here: https://sra-pub-src-2.s3.amazonaws.com/SRR14667226/CTRL_possorted_genome_bam.bam.1

Please get that with curl/wget and convert using the 10x utility.

You need to look for the BAM files under Data Access tab (for above file: https://trace.ncbi.nlm.nih.gov/Traces/?view=run_browser&acc=SRR14667226&display=metadata ). They may not always be available. 10x data submission at SRA can be a hit and miss thing.