Entering edit mode
8.6 years ago
Chen Sun
★
1.1k
I download a SRA file from NCBI, with 166G size.
Then I use fastq-dump -I --split-files ./File.sra to convert it to fastq file on a server with 1Tb memory and 10K RPM disk.
With four hours it only generates 400M fastq file, it seems that convering the file will take forever.
To be precise, its converting speed is about 2M/min, which means it needs around 60 days to convert this file...
Nature Publishing Group journals require that DNA and RNA sequencing data is made available through the SRA(http://www.nature.com/authors/policies/availability.html). It does not make sense to upload SRA format if it is so inconvenient.
Is there a better and faster way to convert, or I am using wrong command?
Check EBI-ENA with the accession number. Get fastq files from them directly without having to mess with SRA data.
Thanks, this idea is very useful!!
What kind of storage system are you using? Some storage systems that are not tuned for a large amount of random access may be problematic (or at least have been in the past).
It is hp SAS disk.
A 166 Gb SRA file is huge! What is the compression ratio for SRA to fastq?
Just out of curiosity, what kind of sequencing is it? Genome, ChIP, RNA?
it's next generation sequencing data, hiseq raw reads for genome