How to submit Illimina fastq to SRA?

xhuister · 09-22-2010, 07:51 AM

Hi all,

Does the anybody have experience in Illumina data submission to the Short Read Archive (http://www.ncbi.nlm.nih.gov/Traces/sra)? After reading official documentation (http://www.ncbi.nlm.nih.gov/Traces/s...Guidelines.pdf), I still don't know how to.
I've already created study, experiment and run through the web interface, but I don't know how to upload my read data and link it to the study or run?

Thank you!

malachig · 09-22-2010, 03:54 PM

If the data is RNA-seq data, then I would recommend creating a GEO submission instead. Then GEO's pipeline automatically creates an SRA submission for you. In my opinion, their instructions and website are more user friendly.

xhuister · 09-23-2010, 05:57 AM

Quote:

Originally Posted by malachig

If the data is RNA-seq data, then I would recommend creating a GEO submission instead. Then GEO's pipeline automatically creates an SRA submission for you. In my opinion, their instructions and website are more user friendly.

Thank you Malachig. My data is not RNA-seq data, but like that. I mailto the SRA yesterday, now I probably know how to do this but haven't tried yet. The rough steps: First create a RUN module through SRA website, then mailto SRA to ask for a FTP password and upload the data to FTP (not compressed), then they will process the data and update the info for your submission.

colindaven · 09-23-2010, 06:09 AM

Hi, I found the european ERA staff at the EBI to be very helpful. They had pretty good documentation too, but you need to email the staff there first who provide you with the relevant files, as the style sheets you can download are confusing.

Good luck

xhuister · 09-23-2010, 08:09 AM

Quote:

Originally Posted by colindaven

Hi, I found the european ERA staff at the EBI to be very helpful. They had pretty good documentation too, but you need to email the staff there first who provide you with the relevant files, as the style sheets you can download are confusing.

Good luck

Thank you colindaven! I emailed just now, hope it works.

And, I also tried upload the data to FTP, and the SRA did process my reads but error ocurred:

data-load: run file problems data inconsistent - length of reads found do not match experiment: ''HWUSI-EAS1571_0012:8:1:1017:20197'' data inconsistent - failed to load data with interface version 1.0
2010-09-23 09:06:02 data-load: run file problems
data-load: run file problems
data inconsistent - length of reads found do not match experiment: ''HWUSI-EAS1571_0012:8:1:1017:20197''
data inconsistent - failed to load data with interface version 1.0

my read is like this:
@HWUSI-EAS1571_0012:8:61:19449:1260#0/1
GAACGTCATGAACGCAGAGTGGCCTTGCTGCCACGATCCACTGAGATTCAGCCCTTTGTCGCTAAGATTCGG
+HWUSI-EAS1571_0012:8:61:19449:1260#0/1
eeffffffeafffffcefffedfff^cdddeeec`ddddddb`d^`Y`a`bcdda__b]b_cL^Y_Y_^^^_

Maybe I wrote the "flowcell" and "lane" wrongly (as attached). What's the flowcell? (I just know the sequence but know little about the sequencing) From the read header (@HWUSI-EAS1571_0012:8:61:19449:1260#0/1), I think lane=8.

Ariani_Andrea · 03-03-2013, 03:22 AM

Quote:

Originally Posted by xhuister

Thank you Malachig. My data is not RNA-seq data, but like that. I mailto the SRA yesterday, now I probably know how to do this but haven't tried yet. The rough steps: First create a RUN module through SRA website, then mailto SRA to ask for a FTP password and upload the data to FTP (not compressed), then they will process the data and update the info for your submission.

Hi xhuister,

so I try to send my reads to the SRA but I have some problem. I make project, sample, experiment and RUN. In the last page of run submission they give me a password and an id for ftp private up-load of my data. I've some problem on up-loading data using FTP. I use ncftp for ubuntu, but it seems it does not works.
Have you some suggestion?

MeganS · 03-03-2013, 10:07 AM

xhuister, check your read lengths. That is what the error message seems to be indicating. You could also try entering the flowcell as "Illumina:@HWUSI-EAS1571_0012". That is the format I used and it worked.

Ariani_Andrea, try regular "ftp". I have not used ncftp, but I know sftp did not work.

mcastro · 04-21-2016, 01:39 AM

Hi all,

I was trying to upload several human exomes (fastq files) to SRA but I have several problems and I couldn't obtain and answer from sra so I would like to ask if anybody had the same errors.
I used the SRA submission portal and when I completed all the submission steps I got the following error for several files:
XXX_2.fastq:39969:31:syntax error, unexpected '@', expecting fqENDLINE

I detected that some of the fastq files were created with a different sequencing software which add an extra @ after the multiplexing index in the read names. I removed all these characters and now I got for some files the error:

ile ouudru8m/XXX.fq.gz is not recognized by SP

It seems a problem with the server remote folder (it seems that creates temporal folders as ouudru8m) but I cannot obtain an answer from the sra team and I don't know how to deal with this error. Someone had this problem?

I used the ascp command line tool in order to upload all the files to he remote sra folder.

Thank you very much

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
fastq-dump on SRA files	harlock0083	Bioinformatics	14	10-18-2018 04:19 AM
SRA to fastq conversion with fastq-dump loses sequences	pcantalupo	Bioinformatics	13	10-08-2015 05:09 PM
Error in coverting SRA to fastq	yixf	RNA Sequencing	26	08-09-2015 05:55 PM
SRA to FASTQ	ramashankar16	Bioinformatics	5	10-26-2011 04:03 AM
How convert multiple .sra files into .fastq in one go?	TuA	Bioinformatics	5	05-27-2011 09:32 AM

09-22-2010, 07:51 AM	#1
xhuister Member Location: China Join Date: Apr 2010 Posts: 41	How to submit Illimina fastq to SRA? Hi all, Does the anybody have experience in Illumina data submission to the Short Read Archive (http://www.ncbi.nlm.nih.gov/Traces/sra)? After reading official documentation (http://www.ncbi.nlm.nih.gov/Traces/s...Guidelines.pdf), I still don't know how to. I've already created study, experiment and run through the web interface, but I don't know how to upload my read data and link it to the study or run? Thank you!

04-21-2016, 01:39 AM	#8
mcastro Junior Member Location: Barcelona Join Date: May 2011 Posts: 7	SRA errors uploading fastq files Hi all, I was trying to upload several human exomes (fastq files) to SRA but I have several problems and I couldn't obtain and answer from sra so I would like to ask if anybody had the same errors. I used the SRA submission portal and when I completed all the submission steps I got the following error for several files: XXX_2.fastq:39969:31:syntax error, unexpected '@', expecting fqENDLINE I detected that some of the fastq files were created with a different sequencing software which add an extra @ after the multiplexing index in the read names. I removed all these characters and now I got for some files the error: ile ouudru8m/XXX.fq.gz is not recognized by SP It seems a problem with the server remote folder (it seems that creates temporal folders as ouudru8m) but I cannot obtain an answer from the sra team and I don't know how to deal with this error. Someone had this problem? I used the ascp command line tool in order to upload all the files to he remote sra folder. Thank you very much

09-22-2010, 03:54 PM	#2
malachig Senior Member Location: WashU Join Date: Aug 2010 Posts: 117	If the data is RNA-seq data, then I would recommend creating a GEO submission instead. Then GEO's pipeline automatically creates an SRA submission for you. In my opinion, their instructions and website are more user friendly.

09-23-2010, 06:09 AM	#4
colindaven Senior Member Location: Germany Join Date: Oct 2008 Posts: 415	Hi, I found the european ERA staff at the EBI to be very helpful. They had pretty good documentation too, but you need to email the staff there first who provide you with the relevant files, as the style sheets you can download are confusing. Good luck

03-03-2013, 10:07 AM	#7
MeganS Member Location: US Join Date: Sep 2010 Posts: 14	xhuister, check your read lengths. That is what the error message seems to be indicating. You could also try entering the flowcell as "Illumina:@HWUSI-EAS1571_0012". That is the format I used and it worked. Ariani_Andrea, try regular "ftp". I have not used ncftp, but I know sftp did not work.