running out of disk space (Solexa)

[fadista]

Hi,

In my group we have a Solexa machine and we are having problems in saving all data generated by a run. Until now I have been using only the fastq files and some report files but nothing else. (I know that public databases like GEO can accept only fastq files: http://www.ncbi.nlm.nih.gov/projects/geo/info/seq.html)

Questions: Is it ok to remove all the other files for good? Do they retain any important information that fastq files cannot give?

All comments are appreciated.
Best regards.

[swbarnes2]

Some people are keeping every scrap of data, but I think that lots of labs are, for space reasons, tossing most of the intermediate files.

The new IPAR set-up automatically tosses the images, I think, unless you set up some pipeline to store them elsewhere. In theory, the intensity files could be run through alterante pipelines, like Alta-cyclic, so there might be a reason to hang onto those, but what are the odds of you actually doing that with most of your data?

I know I'd feel a bit safer with the whole Gerald folder, or maybe keeping the .prb files and the .seq files too, so you can remake the fastq, if you had to.

[Joann]

blobs anyone?

see bloby images at

http://www.genographia.org/portal/topics/example-data-sets/illumina-solexa-examples/blobs-in-images

what others say at

http://flow.yellowcouch.org/wiki/index.php/Image_analysis_and_Base_Calling

Save all data=quality control, unexpected discrepancy resolution between experiments and labs. Resequence vs reanalyze? What if there is no more sample?