Would there be interest in a new site, specific to next-generation-sequencing questions and answers in stackexchange?
It would initially appear in Area 51 to see if it picks up momentum, then have it as a more specific Q&A site than biostars.org. I have seen comments that there is a faction of biostars annoyed by the fact that most of the content in biostars is just a mirror to what is seen in seqanswers, but with a Q&A format.
Having a next-gen-sequencing stackexchange site would solve the problem, and separate into 3 main levels in Q&A format:
I think that SeqAnswers already covers this niche. They already have a large community, and they published a paper about their website in Bioinformatics.
Although stackexchange websites are cool, I don't think it will be useful to have another website on the same topic. In any case, you should ask the same question in SeqAnswers, and see what they think.
I don't see any reasons to split from biostars. The tags are there to help you find interesting questions and filter. if you aren't interested in NGS questions then filter these out. However, more generally there are other types of question where it is useful to have a critical mass of people - I found some of the most useful questions on here about things that aren't actually domain specific. I'm thinking of the hardware purchase questions, journal club gaming and other meta questions. The impact of splitting on such discussions would be deleterious.
I think you have address several questions:
What would have a more specific site help?
What is currently missing from biostars that a stackexchange site would solve?
I don't think there is need for that yet, 60% of questions on Biostars (according to me) are related to NGS, if they are gone, there will be very less traffic to the website and it will make it less exciting to visit and answer.
I don't think a mass migration would be a good idea. We would be losing out on a lot of content that has already been gathered on this site. If the issues are with site design and content management, then we should tackle that issue rather than dump all our existing data for a new system.
Another Q&A site for NGS will be a bad idea. We (here in Biostars) cover vast percentage of NGS areas complementing SeqAnswers. More over too specific areas will lessen the interest for the broader bioinformatics community. I beleive we are good in the current state.