Do you want to run this as an open source project? Also, are you going for standard lab protocols (of say the pharma variety), or reproducible statistical analysis as well?
As for the type of protocol, since the site is oriented around user-moderated groups, we're absolutely down with any kinds of protocol our users are interested in.