Hi,
we sequence human DNA with 454 Junior, single-read. Runs are ok (number and quality of reads).
After trimming (quality>20 or 25 according to tests) and removing adapters (15nt), the reads distribution according to the length is quite a Gauss curve except a peak at the max reads number for a given size. This peak describe reads with the same length for most passed runs (499nt if qual>20, 483nt if qual>25). We don't observe this peak before trimming and these reads (as runs) match with our interest region (no contamination).
Ours questions:
- why is there a peak and not something like a Gauss curve?
- why always the same length (after trimming) regardless runs?
Can we say that the polymerase enzyme and reagents are really as reproducible as possible that, at the cycle n=X, their quality/efficiency falls?
quite a planned obsolescence of biological (and commercial) products
Thanks for your help!
> problem solved by biologist (error with the sonication step...)
we sequence human DNA with 454 Junior, single-read. Runs are ok (number and quality of reads).
After trimming (quality>20 or 25 according to tests) and removing adapters (15nt), the reads distribution according to the length is quite a Gauss curve except a peak at the max reads number for a given size. This peak describe reads with the same length for most passed runs (499nt if qual>20, 483nt if qual>25). We don't observe this peak before trimming and these reads (as runs) match with our interest region (no contamination).
Ours questions:
- why is there a peak and not something like a Gauss curve?
- why always the same length (after trimming) regardless runs?
Can we say that the polymerase enzyme and reagents are really as reproducible as possible that, at the cycle n=X, their quality/efficiency falls?
quite a planned obsolescence of biological (and commercial) products
Thanks for your help!
> problem solved by biologist (error with the sonication step...)