To screen or not to screen

So in my copious spare time, I've been approached to co-write a metagenomics grant. We have the RFA and we have our system (rhizosphere). Now comes the fun part. Sketching out years 1 through 5, experimental design, and budget. If we're successful (we need to get past the "Letter of Intent" stage first), we hope to fund a post-doc to run with this project. I'm thinking that we start off with a good 16S rDNA deep sequencing run or two to establish the major players in our system, and follow that up with a metagenomic sequencing run. I favor the shotgun approach, just sequence everything we come up with and take it from there. I know other folks like the screening approach and only going for things that they think will be beneficial to know about. Off the top of my head, I can't think of an easy screen for our system, and while it'll be a lot of extraneous info to mine (and a more expensive proposition) I think it'll be worth it. From there, I hope to move us into transcriptomics which I think has the benefits of metagenomics while also giving us "real time" data on how the organisms are responding to environmental conditions.

Should be fun. Now I have a couple of days to put my letter of intent together.