Tools For Building A Protocol Repository?

Question

[Anthony Liekens]

In our department with ~100 people, the protocols used in the labs are often hard to find as the knowledge is distributed among too many people. We are currently pondering solutions for centrally organizing this documentation, protocols, laboratory and stock guidelines, introductions for newcomers, etc. We are wondering what solutions we could put into place that would motivate people to commit content in the long term.

Having done a few documentation systems, repositories or wikis in other places, I have realized that starting such a plan always is greeted as a good incentive with a lot of optimism, but motivating people to continually use and improve the content in such systems is something else, because it is not always clear how the committer wins from this.

What solutions (document repository, wikis accessible to non-computer-geeks, proprietary solutions ...?) have you tried and which ones did succeed in getting people to commit to in the long term? Have you cracked the problem to get people motivated to commit time helping the rest of your department? Are there any out of the box solutions, like gamification, that have helped?

[Ying W]

A wiki similar to this? http://openwetware.org/wiki/Protocols

Answer 1

[Damian Kao]

Most protocol management systems I've seen people come up with are basically just a wiki-type content management system. The problem with that is protocols are not static things.

In our lab, there are probably 3-4 versions of the same protocol floating around with minor changes. Some people felt adding more of something makes it better. Others found that certain conditions works best for their specific experiment. And some just prefer reagents from other companies. None of them are "right" or "wrong", just better or worse for various situations.

A protocol system would usually just display the original protocol that all those other versions came out of. That really isn't very useful as students want to see the minor changes people have made.

If you can come up with a protocol versioning control system where a "root" protocol template is stored and users can fork the root protocol and branch off into various versions, then I think that would be really useful. Perhaps even a visual system where people can view the protocol as a graph with each step of the protocol as a node connected to the next step with branches for different versions.

[Dave Bridges]

i understand thats the world, but is it really best practice to encourage 4 people to do experiments 4 different ways?

[Damian Kao]

There is nothing wrong with doing it 4 different ways. No protocol is a catch-all protocol that'll work under all situations. An in stu probe made to mark mRNA transcripts of a certain gene can vary hugely in binding proficiency based on the transcript sequence. In those cases, certain steps will need to be longer or more reagent will need to be added.