Paired end data loaded into geo as single end runs ... How to extraxct the data

Question

[ChIP]

Hi,

This sample run http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSM835232 at geo has the sra files in bit different manner submitted.

Such that the mate pairs are loaded to SRA as single end runs resulting in two files per sample.

My problem is, how can I get proper fastq files from these two SRA files.

I tried

fastq-dump -A SRR364680.sra
fastq-dump -A SRR384964.sra

and after that bowtie, but it doesn't work. Has anybody ever dealt with such a data, if yes how can I proceed to get unaligned FAstq files that can be used for alignment.

Here is the head of the two fastqs

file 1

@SRR364680.sra.1 SFGF-GA2-1_63:2:112:1559:999 length=80
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
+SRR364680.sra.1 SFGF-GA2-1_63:2:112:1559:999 length=80
################################################################################
@SRR364680.sra.2 SFGF-GA2-1_63:2:112:9048:999 length=80
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
+SRR364680.sra.2 SFGF-GA2-1_63:2:112:9048:999 length=80
################################################################################
@SRR364680.sra.3 SFGF-GA2-1_63:2:112:10809:999 length=80
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN

File 2

@SRR384964.sra.1 SFGF-GA2-1_63:2:14:1899:1000 length=80
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
+SRR384964.sra.1 SFGF-GA2-1_63:2:14:1899:1000 length=80
################################################################################
@SRR384964.sra.2 SFGF-GA2-1_63:2:14:11711:999 length=80
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN
+SRR384964.sra.2 SFGF-GA2-1_63:2:14:11711:999 length=80
################################################################################
@SRR384964.sra.3 SFGF-GA2-1_63:2:14:13989:1000 length=80
NNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNNN

Thank you

[GouthamAtla]

Can you post the first 5 read names from two fastq files ? What's the error with bowtie ?

[ChIP]

Please check the updated question.

[GouthamAtla]

I am wondering, if it's paired end data, it will have the read1 and read2 information in the read name (like #1,#2 or /1, /2 etc) to distinguish the read pairs. But I don't see them here. The read pairs (R1 and R2) should be in same order for alignment.

Answer 1

[Madelaine Gogol]

I wonder if this is what you're dealing with? How To Convert Sra-Lite Paired-End Submission To Fastq?

[ChIP]

I don't think it is, since the submitter at GEO submitted and loaded mate pairs to SRA as single end runs resulting in two files per sample. The split utility of fastq-dump is not helping in this case.

Answer 2

[skittely]

Based on the original geo page for the .sra reads, it seems the reads are single end rather than paired. http://www.ncbi.nlm.nih.gov/sra?term=SRX105932 - if you see the Library section and click on "more..." then you can see the layout of the library. My guess is the two separate files are some sort of replicate.