454 run - wrong primer - can it be salvaged?

[Indiana]

Hello,
New here. Hope to find some ideas about something. Just did a 454 run (FLX+) and grabbed the enrichment primer instead of sequencing primer. Result: zero key pass, but images show run was successful. Any way to convince the pipeline to process the wells that did not key pass? Or are those cycles needed for proper calibration? Have looked at this thread, and am wondering if this approach is potentially useful:

http://seqanswers.com/forums/showthread.php?t=15169

Side note: I'm not sure where the enrichment primer binds, so I don't know what was actually sequenced.

Thank you ahead of time,
James

[flxlex]

Could it be that what you see in the images is control beads only?

[Indiana]

Not control beads

No, because the control beads actually worked (they are pre-primed) and they look fine. I'm now worried that since the correct primer was not there, then the first few bases could be non-individual (meaning, it might be TTAAAGGG, or something like that) causing the software to calibrate incorrectly.

[pmiguel]

Sorry, your run is lost. The key bases are absolutely required at an early stage of processing.

Okay, maybe there would be a way. This presumes that the enrichment oligo primes somewhere distal to the insert from where the sequence primer anneals. So your sequence would start inside the adapter but would eventually read out into the insert.

You would need to find out where the enrichment oligo actually primes from. Possibly you could page through the flows with gsRunBrowser until you see where they key bases show up. Then you would need to get rid of all the tiffs prior to that point and convince the run processing software to start with later tiffs. Possibly you could just rename the all the later tiffs to appear like they were from earlier cycles.

But my advice is to cut your losses and do the run again.

--
Phillip

[nicorascovan]

Hi James,

Could you save the run at the end? I made the same mistake and I would like to see whether I can save at least some data from the run.

I hope you found a solution and a way to recover the sequences that started from the enrichement primer.

Thanks,

Nicolas

[Zaag]

No key = no data. It uses it to normalize the signals so everything is lost.

[nicorascovan]

Great!
Thanks for you answer. Roche support told us the same. A thousand bucks to the trash....

[martin2]

Rescuing the 454 run

You can disable the key check in software, it is even described in the Roche manual. Just edit the XML template file for the processing pipeline. Probably you will need to disable all filters but to get something out it should be acceptable. One could later on re-enable some read trimming/discarding steps.

If you send me some raw reads I believe I would have more clues what went wrong or at least what else you could adjust. And, I think I can do the primer/adapter removal using my own tool so that you can really disable that task in Roche pipeline (the first paragraph).