why SRA converted fastq file is like this?

[hujiacun]

Hi,
I converted a ncbi SRA fileto fastq. According the original paper, the read should be pairend, and each end is 100bp. Why the two reads in the fastq are add together, and length is 200bp? How can I split this to two files?

[GenoMax]

You should add "--split-files" option to your fastq-dump command to split the two reads into separate files. May also want to add -F to get the original read headers.

[hujiacun]

Thanks! Can the split command do for this format as below. The two paired end is put together(100+100=200). As no separating gap between the two end, how can fastq-dump pick them out? Any command option is required?
"--split-files" or " --split-spot ", which one is right for me?

Format as :

head -100 /home/hu/mydoc/SRR616966.fastq
@SRR616966.2 FCD0R1VACXX:5:1101:1911:2152 length=200
ATGTATGATATAACAGTAACTATGTGAAACAAAACAAGACCCGATNNAANANNGACTATATTATAACATCATCATATCAACGGTGACTTCAAAATAAAACAGATAGTAGTTGTTTTAGATTTTCATGCCTTAATAGTGTTTTAGATATGAGGCTAATTTCTATGCTATGACGCTGCTATATATAAATTTATAATAGTCCT
+SRR616966.2 FCD0R1VACXX:5:1101:1911:2152 length=200
a[acceeefggggiiighiihiiighhhhfiiiihiihiiiiihiBBOOBOBBMMaefhhhhhiiiiiggggggeeeeeeeca^accccccccccccccc___eeeccgggggiiihifhiiiifiiiiiiihihi`efhiiiighhhihfgheifhihiiihiiiihihiihfghifggfegggedeeeeebacdddcd

[GenoMax]

Use --split-files and -F with the original SRA# or downloaded sra file.