32bp?

[yamayaya]

Hi, do anybody have ideas about "32bp" in solexa system?

In user guider, it says that 32 is the max ELAND_READ_LENGTH and it is a seed for sequence reads longer than 32bp.

So, I am wondering whether it will come the exactly same results even when I involve 36bp for ELAND analysis?

[swbarnes2]

I think with longer reads, it only looks at the first 32 bases, and just tacks on the rest of the read after.

So I think the answer is yes.

[dcjamison]

The first 32 bases are used for the seed alignment, then the alignment is extended for the bases beyond 32.

For GA Pipeline 0.3 and newer, there is a Gerald analysis mode called eland_extended which takes the top potential alignments and extends all of them, scores the alignments with a phred-like score, then reports back the best scoring alignment. So, while only the first 32 bases are used for the seed alignment, all bases are used to determine the best match.

[yamayaya]

Thank you very much for kind help.

Yes, we have GA pipeline 1.01 on hand. So you mean, if we use eland for 32bp analysis, it will only look at first 32bp, but if eland_extended was employed, it will involve all 36bp for inspecting the score?

[dcjamison]

Yes, that is correct. eland_extended also produces better output -- the *_export.txt file is much easier to use (in my opinion).

[inesdesantiago]

Hello,
Do you know if there is a way to use the alignement scores of the _export.txt file to plot statistics on quality of the alignments?
I mean, what are this scores good for? What to they mean?