Biology Stack Exchange is a question and answer site for biology researchers, academics, and students. It's 100% free, no registration required.

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I am aligning a large number of ESTs. It seems poly-A tails show in many different ways. In addition to occurring at the very end, they can be flanked by the cloning sequence one one end, or have mismatches/errors. What is a good rule or available tools that will handle the usual cases?

A few examples of the non-trivial cases I found, with their Genbank Accs:


Cross-posted on SeqAnswers,Biostars.

share|improve this question
Did the answer on Biostars not work for you? (If so, what is it that it doesn't do that you would like it to do?) – dd3 Jul 10 '13 at 20:22
@dd3: trimest has an option for trimming the poly-A regions, but we have to specify the rules. I have already done the basic filtering, but what are the "good" rules? Is there something better than min-length and max-mismatches that will handle real sequences I have given examples of? – highBandWidth Jul 10 '13 at 20:34
You could try the empirical approach. Take a random sample of N ESTs and calculate the distributions for min-length and max-mismatches. Use the distributions to pick appropriate cutoffs (some sort of statistics thing). – dd3 Jul 10 '13 at 20:55
@dd3, that definitely makes sense and seems like a good idea. (although there's some judgement to be made for things like ...GAGAACTGGGCAAAA: where does the tail start?). Since EST processing is well established, shouldn't something like this already exist in the tools or the literature? How does everyone else do it? – highBandWidth Jul 11 '13 at 3:09
Use dust and remove all low complexity sequences.. Or simply use a regex search to filter out poly-As – WYSIWYG Aug 22 '14 at 6:07

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

Browse other questions tagged or ask your own question.