The times I have sent a sample to sequence, both the forward and the reverse primer sites, show high inaccuracy while the rest of the gene is correctly sequenced. Because of this, the sequences of my in silico construct and the sequenced sample do not align in this section; but they do align at the rest of the gene almost 100%.

Is there a reason for this? Is this simply a sequencing artifact or is should I trust the sequenced sample and assume that the primer sites have mutated?

  • 1
    $\begingroup$ The beginning is always garbage, you can safely disregard. Though, the fact that you have a sequencing result at all means the 3' end of your primer matches the template. $\endgroup$ – Eliane B. Sep 12 '18 at 10:02

The very extremities of sequencing reads obtained by most if not all sequencing technologies are usually of lower quality, though more often so in the 5' region. You should disregard this data, or better yet design additional primers further away to encapsulate that region too if you desperately need it.

Below is a fairly typical output from FASTQC analysis of Illumina sequencing data. You can see how the quality is at its peak in the middle of the read (base index on the x axis. You'd likely see a similar thing for Sanger sequencing which is presumably what you're using.

enter image description here

| improve this answer | |
  • $\begingroup$ What would a good score mean? and what would a good score be? I imagine from the colors that anything on green is good, but just to know a bit more about sequencing. $\endgroup$ – Franco Grosso Sep 14 '18 at 10:13
  • 1
    $\begingroup$ It can vary by sequencing technology, but most techs these days use something called “Phred-33” scoring. It’s a logarithmic relationship that roughly speaking means a Phred score of 30 (on the Y axis in the pic above) means there’s only a 1 in 1000 chance that base was called incorrectly. Most people accept 30 as a reasonable quality for sequencing, also helped by the fact that you have many millions of overlapping reads (in NGS), so even if a base is miscalled in 1 read, chances are there are many other reads you can refer to that would have the correct call. $\endgroup$ – Joe Healey Sep 14 '18 at 10:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.