So … what goes around comes around. In 2003 and 2004, I spent a lot of time discussing and arguing with people about what would be the best strategy for making and sequencing Sanger libraries for metagenomic sequencing for the Sargasso Sea metagenome study coordinated by the Venter Institute (I worked at TIGR at the time and was collaborating with people at the Venter Institute on this project). Then, a short time later, we had similar arguments and discussions for the Global Ocean Sampling (GOS) project (again, with Sanger sequencing). Then as next gen sequencing came online, the argument changed a bit (since it was not quite as simple to make mate pair or paired end libraries or sequence them). And every year or two I have been involved in the same type of discussions. Without wanting to prejudice people’s ideas about this, I thought it would be a good time to solicit input from the community on metagenomic sequencing and library construction.
So – I would like to solict such input with some questions:
- What sequencing method(s) are people using for metagenomic studies?
- What are your approaches to library construction (method, size of DNA pieces, etc)?
- What are your approaches to sequencing (paired end? read length? overlaps or no overlaps?)
- Do you use one approach or many combined together?
- How did you / do you choose what approach to use?
- How much do the goals of a study (e.g., assembling genomes, classifying organisms, comparative ecological analyses, functional discovery, etc) influence the choice of library and sequencing methods
- What papers, blogs, data sets, Tweets, etc have you found most useful in making such decisions?
Some discussion of this post on Twitter