How can i get a merge fastq file from sratoolkit like the file i directly download from ncbi?

Question

[Little.G]

Excuse me, everybody. I am a new one here and also this field. I need help a lot .T-T

Parts of my story... I downloaded SRR__.sra file from ncbi using sra toolkit. i used fastq-dump to convert SRR__.sra to fastq file and splitted file into 3 files : 1.SRR____1.fastq 2.SRR____2.fastq 3.SRR___.fastq which is the combined version of 1+2

My question are.... 1) If i don't misunderstand, SRR___1.fastq and SRR___2.fastq are from the paired-end sequencing,right?

2)How can i merge the two file together? i want a single file to do the downstream process like mapping etc.

3)How can i get a file as a directly downloading file from ncbi website using sratoolkit? Was the directly downloading file already merged? (In the file, every sequence has optional description after seq name)

Please give me some suggestions. Thank you you guys in advance!

Answer 1

[h.mon]

Download the fastq files directly from ENA, it is simpler, less error-prone and much faster then fastq-dump.

You don't need to merge the forward and reverse reads, all mapping programs can take the two files separately.

[Little.G]

Thank you so much. It's really faster !!

Finally,If i want to analyze for DEGs, is there a step should i combine the two reads together? I used to practice on a single read (in my class). I compared gene expression in rice planting in two conditions(triplicate each).

[h.mon]

I will say it again: no, you don't need to merge / combine forward and reverse reads (or R1 and R2), mapping programs (such as STAR, HISAT2 or Subread) take both R1 and R2 files as arguments, and take into account the paired nature of the reads.

Answer 2

[mmfansler]

1. Yes.
2. Don't do that. Instead, use the --split-spot flag to extract an interleaved FASTQ. E.g., fastq-dump --split-spot SRR390728.sra

3. fastq-dump will automatically download from the Sequence Read Archive if you provide the accession instead of a local file. So, in the end you want:

   fastq-dump --split-spot SRR390728

More details in the manual.