Crowdsourcing a response to an NIH grant opportunity

UPDATE: This opportunity has passed

A tremendous amount of disparate biomedical research data is created every day. Wouldn't it be nice if all that data was interoperable, openly licensed, and packaged ready for researchers to gain insights from?

The National Center for Advancing Translational Sciences (NCATS) seems to think so! They're providing $5 million in funding for 2-5 groups to lay the foundation for the creation of a "comprehensive biomedical data translator".

A Thinklab proposal to crowdsource it

NCATS says the translator project will "require unprecedentedly broad teams of experts to work together in a highly collaborative manner". While many traditional academic groups will be applying, they're also open to innovative responses from non-traditional sources such as informal groups from within the open source community.

This is exactly this kind of opportunity I had in mind when I created Thinklab. The purpose of this post is to find people with relevant expertise who are interested in participating in a Thinklab proposal.

Here's how we might distribute $1 million in funding:

• $500k in salary to core team members. We're going to need a core team of researchers who are prepared to commit at least 25% of their time to this project over the next 1-2 years. Our key differentiator is this: everything will be done in the open. A great example of this is Daniel Himmelstein's Thinklab project: Rephetio: Repurposing drugs on a hetnet. As it happens, Daniel's work is highly relevant to NCATS's translator project. Check it out!

• $300k in payments to online contributors. In order to encourage the most broad participation possible, a separate pool of funds will be set aside to compensate contributions from the community. Anyone who has relevant expertise (including non-U.S. based people) can share feedback or make contributions and be financially rewarded for their efforts. Earnings will be entirely based on peer assessment of value added. Thinklab recently paid out $2,557 under this exact model.

• $200k in small grants for data analysis projects. As a result of this project we expect to generate many open integrated biomedical data sets. To get the community involved and have real use cases to work with, we will award small grants for data analysis projects. More ambitious applicants may choose to integrate novel data sets and base their analysis on that. There could also be room for grants to develop various data interfaces.

The above highlights some of the unique aspects of a Thinklab proposal. The actual proposal will detail more of what we'll actually do. For right now I just want to get you interested.

Here's the pitch: by participating in this project you'll have an opportunity to build a foundation that will catalyze much of the future of biomedical research. Not only that, but you'll earn real money (and recognition) based on peer assessment of the value of your contributions. We're exploring a whole new way to conduct and reward academic science!

Now, I need to make a confession: we only have 5 days to get the proposal in. Yikes! But if there's enough support we will make this happen! I need your help though. First, if you're coming from Hacker News, please vote this post up!

For further reading check out NCATS's data translator FAQ or Thinklab's introductory blog post.

Thanks!