Repository/User Libraries/User Input

User comments:

A list of some rough ideas follows, I may have more to add later. Best, Julie Norville

1. The importance of a web based registry/library to store individual physical instances of parts at the lab level

Based on my experience, I think it would be useful if the library system is something that you could access on the web (i.e., part of the registry) rather than something that is stored on an individual computer. I am not sure whether or not you would want all of the items in such a library system to be open to the public, before their release to the shared registry (though perhaps they should be open within a lab for individual lab accounts)-- though it does make sense for iGEM team accounts to be open from the beginning and there may even be reasons why you would want their library or a registry of their freezer stocks to be open. (In the long term it may be advantageous for all labs to have open freezers or registries of their freezers, but I think the culture of the biological world will have to evolve a bit for that to happen. At some point the advantages of a registry system may be so great that individuals will convert to it anyway, but I think faster adoption will occur in the short term if it starts in a semi-closed state.)

2. From physical instances of parts to tracking a part through its growth and development

Why a web system rather than an individual computer system? Information is easily lost if it is stored on an individual computer. It makes sense for there to be sharing throughout the libraries (or individual lab freezers) within a lab, especially when people leave. Thus if the information is available on the web (and stored at the lab account level) there can be continuity as people leave and come to the lab. It may also, make it easier for people to replicate and and take items with them when they leave a lab (if they can change status from lab member to lab leader and have a copy of their virtual parts come with them.) A lab level registry can also help prevent needless replication of standard parts within a lab and help the individuals store their data and plan experiments better. As new tools are developed for verifying parts then everyone can use them to verify their individual parts or collections of parts. The same thing can be done for planning the construction of part, as tools for that evolve and develop. It may also be useful to develop tools that track instances of a part through their evolution-- dna stock, -80 stock, mutagenesis clones before and after sequence verification. If iGEM teams become involved with the development of such tools then the registry can evolve much faster.

3. Eventually the registry can have a reputation verified protocol system built onto the ability to store specific instances of parts, and perhaps support itself through advertisements or sponsored protocols from companies (that are verified by registered users) (As a sidenote, this is starting to look like a company, but as Paul Graham states, many great companies are very similar to great charities when they begin. The expanded registry, if it produces revenue, could be envisioned as a nonprofit whose goal is to make science better and easier for the user.)

Having a web based system will require large storage capabilities, which may eventually incur large costs. It may be possible to support such a system (that tracks individual experiments and their products) with advertisements from Invitrogen, NEB, and other companies. If the registry system evolves to track and help plan experiments then companies may help provide tools to navigate or debug a series of experiments, or even want to market new products to a very targeted user group through advertisements. It is possible that individuals can rank their experience using a protocol every time with some sort of star system, and help let other users know what its best. OpenWetWare provides a feedstock of protocols. iGEM teams could perhaps create a system that allowed you to create tools for performing and tracking the physical products of experiments (something like a "wiki-like" version of a tool-creating-system that is easy to modify at the "top gui level" with a command interface) in the registry to track individual experiments. Once the tool works well, you could release it to the registry or write it up as a BioBrick Tool Note.

This website has an example of a reputation based interface: http://panjiva.com/sample/complete. Alexey Radul at MIT may be good reference to talk to about setting up a reputation based system for the expanded registry when it needs one.

Is this where I add feedback about the new registry design? On the new part page, I cannot find any information about what vector backbone the part is in! This is crucial info, needed for antibiotic resistance info as well as size of plasmid on digests. I appreciate reorganization, but crucial data is now missing!