Difficulty aligning to reference with AVA

[cog]

Hi,

I'm using the Roche Amplicon Variant Analyzer software to align amplicon reads from GS Junior to a reference sequence. The reads have already been successfully manipulated (split by tags) using SFF Tools.

My problem is, after performing the computations with AVA I get the message "No sample/reference data to align".

I've set up my project following the instructions in the Roche "Part D" manual, double checked all my associations, and the project structure looks similar to the ones in the examples. I've worked through the instructions a couple of times and can't seem to find where I'm going wrong. Any ideas?

Thanks!

[kmcarr]

Cog,

Are you specifying MID information when you set up the mapping project? If the sff files have already been split by MID tag then you should NOT specify any MID information when you use these with the Roche assembly/mapping tools.

Splitting the sff files also sets the clipping information to remove the MID tag from the 5' end of the read. If you subsequently tell the mapper that the reads are MID tagged it will look for the tag but not be able to find it because the tag has been clipped. The mapper will thus reject all reads at the filter/indexing step because none of them appear to have the specified MID tag.

[cog]

Hi kmcarr,

Thank you for your reply. You are right, the files have already been split by MID tags. I checked my project in AVA, and there were indeed MIDs defined by default. I removed these (from both the project tree and the definition table - just to be sure!), saved the project, and ran the computations again, but I still get the same error.

Are there other filtering processes that might be rejecting my reads?

Many thanks for the help!

[menenuh]

I have the same problem too. Any suggestions?

[cog]

Hi Menenuh,

A part of the problem might be because I split by tags using SFFTools, before going into AVA. If I go back a step to my original file, and instead use AVA to identify the tags (by defining the MIDs in AVA directly using the "Multiplexers" tab, and then defining all the samples and amplicons seperately) it worked better.

But it's not a complete solution because I found that AVA's demultiplexing had successfully assigned only small numbers of reads to each sample (<30 reads). From using SFFTools I know that I should have 100's of reads per sample that have perfect matches to my MIDs.

I would be interested to hear if you have the same result with your data.

-cog

[menenuh]

Hi Cog,
I have also used sfffile command to split my reads before going to AVA, then bumped into same "No sample/reference data to align" error. I haven't tried to use the original sff in AVA, but I would be hesitant to use it if you say it's not specific enough in demultiplexing. Besides, I have 48 MID tags, that would take some time to import

[westerman]

I have a Amplicon project on-going and am running into the same problem -- e.g., everything seems to be set up properly but am getting the 'no sample/reference data to align' message. Did anyone ever get another clue about this? Thanks.

[kammoji]

No sample/reference data to align

Hi all,

Same problem here. Everything apparently is set up as in the manual example. Already tried removing the default MIDs. No effect.

I have the sample associated with the reference. It is visible right below the reference, and I have only a single reference, 1 sample and 1 amplicon. How can it be that the program cannot find it?