Split fastq.gz file

Question

[sandy]

Hello everyone,

I download a single fastq.gz file from a study posted on GEO, and this is a paired read single cell sequencing. Somehow this study compressed R1 and R2 file into one fastq.gz file, I used zcat to check the reads and they are really long, like 98 characters. I want to split them to a R1_fastq.gz and R2_fastq.gz files. Please give me some advice on what should I do.

[GenoMax]

Can you post the accession number so we can examine the record? From where and how did you download the file?

[sandy]

Hi, thanks for your replay. I download this study: GSE138669 from GEO. I used download link generated by SRA explorer using ascp download the fastq.gz.

[ATpoint]

Please show a head.

[sandy]

sure!

This is:

zcat SRR10254552_GSM4115872_SC18_Homo_sapiens_RNA-Seq.fastq.gz|head -10 
@SRR10254552.1 NS500211:194:HNFHHBGXY:4:21402:13612:10917
CACCGCGAGGGCGGAGCTGCGTTGTCCTCTGCACAGATTTCGGTGGTACTCTGAAGGCGGAGCACAGTTCTCCTCAGGTCAGACCCGGGCGGGCGGGC

Answer 1

[GenoMax]

Looks like this is a 10x dataset which can be a hit-or-miss proposition to download intact from SRA. Your best bet is to either use the count file provided in the GEO record or use Data Access tab to download the bam format file that has original data https://trace.ncbi.nlm.nih.gov/Traces/?view=run_browser&acc=SRR10254549&display=data-access . Then use bam2fastq utility provided by 10x to reconstruct the original fastq data.

[sandy]

Thank you!

[sandy]

Sorry to bother you again. To download bam format file, can you give me some suggestions about how to download it? Should I use wget, prefetch, or ascp?

Thank you

[GenoMax]

You can use wget with the link https://sra-pub-src-2.s3.amazonaws.com/SRR10254549/SC2possorted_genome_bam.bam.1

[sandy]

Thank you!