Pipeline Illumina question

[bioinfogirl]

Hi,

I'm a new member and I have some questions about Illumina Pipeline GAII.

In Ilumina pipeline documentation there is no documentation about script used by the pipeline. Output of the pipeline show that Goat call different scripts. Does anyone have the details of call script process? What do each script? What is output of each script?...

When I run GERALD as a standalone program with parallelization, I have warning message like 'Warning: did not find any sequences to process (raw)' or 'Warning: none of 25092 data sequences passed filter criteria '(($F[5]>=0.6))' at /env/borneo/pipeline/GAPipeline-1.0/Gerald/qualityFilter.pl line 231.' Does anyone deals with this warning? In this output the lane_tile process are not in order. Does anyone know why? Is it due to the parallelization? This is a problem for me to determine what is the lane_tile concern by this warning.

Last, I would like understand ELAND algorithm. But I didn't find any documentation. Somebody know how can I get this information? saw in documentation that ELAND make several pass alignements. I would like more explanations about the alignment process : pass, error rate...

Thanks

[ScottC]

Hi,

Maybe it would be useful if you gave the commands you are using to initiate the programs.

For a description of the pipeline and how it works, including ELAND, the documents at Genographia.org (mostly written by Nava Whitford) will probably come in handy:

http://www.genographia.org/portal/topics/analysis-methods-and-algorithms

http://www.genographia.org/portal/topics/reports/pipeline.pdf/view

I thought there was a document on there somewhere about how ELAND works, but I don't seem to be able to find it at the moment. I'm sure someone else will post.

Cheers,

Scott.

[bioinfogirl]

Thank you ScottC! genographia give a good description for image analysis but I found no algorithm description about ELAND.

"Maybe it would be useful if you gave the commands you are using to initiate the programs."

I try goat script with default parameter on a run test phiX. Our run is not good we have poor signal intensities. But I'm just starting with GAII and I don't know how deal with this error message and especially how to identify the tile and lane concerned by error in output with the fact that lane and tile are not in order in the output.
If someone deals with this situation, It would be great to share his experience.

[new300]

Warning: none of 25092 data sequences passed filter criteria '(($F[5]>=0.6))' at /env/borneo/pipeline/GAPipeline-1.0/Gerald/qualityFilter.pl line 231

This error means that none of the reads passed "purity filtering". Purity filtering is used to discard reads which are bright in more than one channel (say A and C) and which therefore can not be base called reliably. I believe you can get the purity threshold when running Gerald, but I've not done this for sometime.

So the error is basically tell you "the analysis software doesn't think there are any good reads here so your not going to get any reads from this tile". If this is a GA2 flowcell then 25000 is an abnormally low number of reads anyway and something went wrong. You could post the intensity plots (Linked from Summary.htm, if they were produced) or take a look at the images that came off the device.

[bioinfogirl]

So the error is basically tell you "the analysis software doesn't think there are any good reads here so your not going to get any reads from this tile". If this is a GA2 flowcell then 25000 is an abnormally low number of reads anyway and something went wrong. You could post the intensity plots (Linked from Summary.htm, if they were produced) or take a look at the images that came off the device.

Thank you new300!
Effectively the run have low intensities! This explain certainly this error message.
Do you observe desorder of tile and lane in the output pipeline?