OpenWetWare:Steering committee/NSF BDI Grant/Draft/Old Version

'''The grant has been submitted. Thanks to all of you that helped with grant and letter writing. [[Media:OWWv17.pdf|Here]] is the submitted version of the Project Description. More to come including references and front matter.'''

[[Media:BDI draft.doc|Word Version of Grant]]: Taken from this version on the wiki.

Older version

=Project summary= The process of research in biology requires the generation, maintenance, and dissemination of many types of biological knowledge. Unfortunately, much of this knowledge is stored offline in laboratory notebooks or communicated only directly between researchers. OpenWetWare is a wiki (software making it easy to create, link, and share webpages) that provides a forum to capture knowledge currently unavailable in existing electronic data repositories. Furthermore, OpenWetWare is a collection of communities (from individual labs to multi-institutional collaborations) that provide incentive for contribution of information. Within a year, OpenWetWare has grown from a single lab’s resource to one shared by over 60 labs and 1000 users worldwide.

Further funding will enable OpenWetWare to mature into an essential tool for biological researchers in their day-to-day research. OpenWetWare will be a convenient mechanism for exchanging information and troubleshooting experiments with others researchers. By providing this service for the scientific community, OpenWetWare will be in a position to collect, curate, and disseminate much of the information that would otherwise remain undigitized. To accomplish this goal, OpenWetWare's specific aims are: (1) to provide a forum and tools for digitizing information integral to the process of research such as protocols, equipment, and materials, (2) to release a standardized wiki distribution for researchers, providing an easy to use and flexible method for developing online communities within OpenWetWare, and encouraging communities outside OpenWetWare to generate interoperable information, (3) to foster new approaches to education by both involving students in generation of course content and enabling easy reuse of course materials. Each of these objectives will be achieved by two general approaches, (1) lowering barriers to contribution with templates, tutorials, and help pages and (2) improving the utility of contributed information by developing software tools that enable easier organization and annotation of the existing information.

Intellectual merit
OpenWetWare will allow for the collection, organization, and dissemination of technical knowledge that is currently not stored in an electronic form. This addresses a problem stretching across all fields of biology, and will increase the overall pace of scientific research. Wiki technology also enables tracking of the contributions of researchers based upon edits and community participation, and will provide new opportunities for reevaluating scientific contribution, merit, and impact. We also expect increased sharing and partial standardization encouraged by reuse of protocols will have downstream effects in making data management easier by standardizing the experiments generating the data. Finally, previous success in founding OpenWetWare and fostering its rapid growth suggests that the leadership team is well positioned to succeed in achieving the goals outlined above.

Broader impacts
OpenWetWare will provide new mechanisms for collaboration in research and education. The OpenWetWare-standard wiki distribution will freely provide technical infrastructure for improving collaboration to all scientific communities. The shared backbone provided by the distribution will enable innovations developed by one community to be quickly shared with all others. OpenWetWare provides unique opportunities for blending research and education. Courses have already experienced success on OpenWetWare, suggesting that providing students with increased control over annotating and editing course content is a powerful mechanism for engaging them in the material. By dramatically lowering barriers to contribution we are creating a free resource that is accessible to any researcher with access to the Internet. Underrepresented groups that might face barriers at many levels in sharing information or publicizing their work are able to do so as easily as any other group on OpenWetWare. Additionally, community members have begun creating non-English versions of particular sections of the site, as a means to provide more direct outreach. Lastly, all content on OpenWetWare is licensed under a Creative Commons license ensuring free reuse of the information at no cost to site visitors.

=Project description= OpenWetWare is a collaborative website that provides researchers with an online venue for easily storing, organizing, and sharing information. By encouraging the growth of online communities of researchers, OpenWetWare hopes to capture information that is typically not widely disseminated (or stored electronically at all) such as detailed protocols, equipment usage, reagent details, and much more. To accomplish this goal we are asking for funding to construct a small team to provide administrative and technical support to the existing dedicated and energetic volunteer community. The team will be charged with first making templates and tutorials to increase community involvement by making information contribution easier, and second to develop tools to simplify information management and aggregation. The grant will give an overall motivation and significance (Section B), provide details on how proposed work will further each of three aims (Section C), and finally discusses overall project organization and logistics (Section D).

Overall motivations and significance
The process of biological research generates and uses a wide variety of information, much of which is either inaccessible or unrecorded. For example, consider a researcher who constructs a knock-out mouse of a particular gene. If the experiments are successful and sufficiently interesting, the information is shared to the broader community through scientific publication. The publication would describe the hypotheses tested and the phenotypes observed, but miss much of the process by which the researcher did the work. Information such as how primers were selected, PCR optimizations, experiences making the transgenic mouse, and failed approaches are unlikely to be disseminated. Even worse, if the project is unsuccessful or uninteresting, any useful information generated along the way is not stored or broadly communicated.

Traditionally, much of this information is stored offline by the researcher in their laboratory notebook, and as a result the information is inaccessible to the wider community. Even within a laboratory, much of the information is lost when the researcher departs; and the information that is passed on usually takes the form of methods, procedures, and tricks that lose the original context on how they were developed. This information is often passed within a laboratory for many scientific generations, allowing individual laboratories to become specialists at a particular set of methods, with little information transfer outside that laboratory. This particular example could be generalized not only to biological hypothesis generation and testing, but to computational and experimental methods development, equipment and software usage, and even educational coursework development.

The ability to track and store these types of information, which provide the contexts, experiences, and methodologies of research would be useful for scientific progress. We often cannot learn from others' mistakes, because failed experiments are not published (ref). Laboratories often struggle to repeat the others' work due insufficient information (ref). Lack of collected experiences on commonly used protocols prevents avoidance of common pitfalls, interpretation of failed experiments, and development of new, more reliable methods. Much of this information is not shared because there are neither many incentives, nor easy way to digitize, store, and organize such information. Furthermore, researchers that are sharing information are doing so offline, or in an ad-hoc manner that does not make the information easily available to a wider community. Finally, hypothesis based scientific publication is often inappropriate for publishing such information because of space restraints, lack of direct importance to the results, and because inundating readers with too many details detracts from the publication's message.

OpenWetWare is a wiki dedicated to capturing and curating the day-to-day knowledge of researchers at the bench that is otherwise lost in offline lab notebooks or shared only in small communities. A wiki is a piece of software that provides tools to allow many people to easily and effectively generate, edit, and link between content Cunningham-2001. OpenWetWare started approximately one year ago by graduate students in our lab as a tool to digitize, store, and share this lost information. Their hypothesis was that using a wiki to democratize content generation would provide a powerful tool for sharing information within the lab and with our close collaborators. As our collaborators began also joining the wiki, OpenWetWare's ability to foster collaborative projects and information resource development became more clear. We soon opened the site up to other labs at MIT, and then more broadly as an open invitation to the scientific community. As of June 2006, the wiki has over 1000 individuals and 60 research laboratories who have contributed to a site with several hundred edits and 40,000 page views per day. The users have generated many types of useful information that cannot be found elsewhere, such as: up-to-date individual and laboratory research directions; protocols, notes and tricks, and/or expected results on hundreds of biological procedures; laboratory notebooks detailing ongoing experiments; information on equipment operation, calibration, and control experiments; aggregated informational resources on strains and genotype information; collaborative project discussions and data; community generated information portals on particular fields; safety information and procedures; etc. In addition, professors at MIT and other institutions successfully developed and taught courses from sites developed on OpenWetWare, demonstrating the success of the wiki in an educational environment.

This proposal seeks to have OpenWetWare continue to serve as a way for individuals and groups to store, organize, and disseminate information that serves themselves, their scientific communities, and the world at large. Funding would provide a dedicated administrative and technical team to compliment the existing community based organization to make OpenWetWare an easier information generation and collaboration tool, and also provide new mechanisms for information management and organization as more types of data come online. We expect that continued success of OpenWetWare and its goals will have many large impacts on biology. First, as a direct of result of easy and democratized contribution, the scope and pace of scientific communication will increase dramatically. Second, the accountability of editing and breadth of information about researchers will provide new opportunities for reevaluating scientific contribution, merit, and impact. Third, biological course development, student involvement and feedback during courses, and reuse of educational materials will improve dramatically due to the new tools for easy content development and information aggregation. Finally, OpenWetWare will provide a new understanding for the process of scientific discovery and the current state of research for the funding agencies, ethics groups, and the general public.

Research plan
Further funding will enable OpenWetWare to mature into an essential tool for biological researchers in their day-to-day research. OpenWetWare will be a convenient mechanism for exchanging information and troubleshooting experiments with others researchers. By providing this service for the scientific community, OpenWetWare will be in a position to collect, curate, and disseminate much of the information that would otherwise remain undigitized. To accomplish this goal, OpenWetWare's specific aims are: (1) to provide a forum and tools for digitizing information integral in the process of research such as protocols, equipment, and materials, (2) to provide an easy to use and flexible method for developing online communities within OpenWetWare and to encourage communities outside OpenWetWare to generate interoperable information by releasing a standardized wiki distribution for researchers, (3) to foster new approaches to education by both involving students in generation of course content and enabling easy reuse of course materials. Each of these objectives will be achieved by two general approaches, (1) lowering barriers to contribution with templates, tutorials, and help pages and (2) improving the utility of contributed information by developing software tools that enable easier organization and annotation of the existing information.

Overview
OpenWetWare provides a resource for digitizing information integral in the process of research. Already, the site has been a useful tool in disseminating information not often found in existing sources. Two key priorities drive the work proposed in this aim. First, improving the ease of use of OpenWetWare so that information generation is easier. Second, developing methods for the automatic organization, annotation, and processing of this information to enhance usability.

Significance
The process of research and discovery in biology requires many types of information. Some, but not all of that information is accessible in the form of scientific publication. Additional information that is often not available in publications include details of experimental methods, tips and expected results on protocols, information on equipment usage, written descriptions of strain histories and phenotypes, current directions of a researcher’s work, et cetera. Hindered ability to access the sum of these details can lead to problems in verifying, utilizing, and extending others’ research.

There are many indications that the sharing of information outside of standard publication materials is currently a problem. For example, in a suite of large-scale studies of sharing in the field of genetics and other biological disciplines, a group of researchers indicate that data sharing in biology is a significant problem Blumenthal-JAMA-1997, Campbell-JAMA-2002, Blumenthal-AcadMed-2006, Vogeli-AcadMed-2006. Many types of information are withheld, from sequence information (28%), pertinent findings (25%), phenotypic information (22%), and information regarding laboratory techniques not used in publication (16%). The study also showed that most thought such withholding of information and materials “slowed the rate of progress in their field of science (73%) and had “adverse effects on their own research” (58%). Eighty percent of those researchers that withheld information responded that a reason for their withholding was that the effort to produce post-publication information or materials was too great. This evidence indicates that easy ways to communicate information outside of formal publication mechanisms would be beneficial.

Most current efforts at storing, digitizing, and organizing information in biology focus on already published information. Often these efforts involve the management and organization of large-scale data sets (refs). Also, efforts have been made to computationally mine the research literature for biologically relevent relationships that may have been missed (refs). These efforts are important, but miss much of the problem with generating and understanding such data. Individual tight knit communities have tried to construct informational resources to capture some of this missed information. For example, WormBase, a resource for the C.elegans community, is an effort to collect much more information than just existing literature and data sets. WormBase collects varied data sets, integrates them across the different genes, conditions, and strains, and compiles other relevant information such as publications and investigators working on the particular areas. The database even compiles information about people, such as where an investigator was trained, and who else that investigator has trained, and publications including conference proceedings and other hard to find information. However in all these cases, the fixed structures of these types of data sets often precludes easy inclusion of new types of information. For example, even WormBase, probably the most comprehensive database on an organism, does not include a repository for laboratory protocols.

OpenWetWare at heart is an effort to digitally store, share, organize, and contextualize such unrecorded information. OWW has excelled largely due to two reasons. First, the open-ended wiki structure allows an expanding array of information to be gathered, interpreted, and most importantly shared. Second, the ease of use of the wiki allows most researchers to not be intimidated with information entry and access. The current challenges we want to address are dynamically organizing and contextualizing the deposited information all the while keeping information easy to deposit. The development and growth of OpenWetWare will provide a capture mechanism for information traditionally lost making it available to the research community at large.

Preliminary results
OpenWetWare began as a modest effort for individual laboratory members to manage their information resources in a simple, open-ended manner. Quickly, individual researchers began adding information that was otherwise difficult to find in a public, electronic form. The scope of such information includes protocols from individuals, unpublished results from laboratories, spontaneous collaborations between individuals solving technical problems, operation and calibration of equipment; The list goes on. Here we will focus in on three specific cases and discuss the types of information being generated.

1. Protocols

Biological research relies on developing and executing experimental protocols and operating equipment. Currently, a researcher usually learns experimental techniques from more senior laboratory members, and then tailors them to her specific research. In the case where no local expertise exists on a particular protocol, the researcher can turn to the scientific literature, where two sources are often available: the methods section from hypothesis driven papers, and papers specifically about protocols in specialized journals or compilations. However, these sources are often insufficient to replicate the protocols for a variety of reasons (ref). First, printed articles quickly become out of date due to ever evolving techniques in biology. Second, often due to page limits, protocols tend to be succinct, and do not give details that are often vital for successful implementation. Third, there is very little feedback on expected data, variation, and difficulty of any given protocol.

OpenWetWare provides a simple system that enables rapid communication of new protocols, unlimited space for description and discussion, allows posting of expected results and common pitfalls, and encourages the communication of the tips that are often important to getting protocols working properly in the lab. The protocol collection on OpenWetWare has hundreds of protocols in different areas of research. Users on the site have found it useful to post protocols because it improves their ability to store, share, and improve their protocols. However, as many people contribute, the combined information becomes a rich resource for information on method development, debugging, and collected knowledge.

Two key trends are worth highlighting because these types of informational pages are rare in traditional sources, but are increasingly common on OpenWetWare.

a) Protocol Aggregation: Several labs posted protocols for DNA ligation using different protocols DNALigation .  Some members of those labs began a more general page describing the general steps of DNA ligations and links to individual laboratory protocols with a description of their differences.  Finally, other individuals added tips, observations, and references to the general process of DNA ligation.

b) User Feedback: A user of OWW posted a particularly detailed protocol on a method for quantifying proteins using an improved ß-galactoside assay BetaGalAssay . The protocol, when initially entered, contained much of the background information on how the protocol was developed, and additional tips and tricks to watch out for.  Another user, who used the protocol from the wiki, then posted her general experiences with the protocol, and sample data demonstrating the repeatability, and general levels of output to expect on a standard control experiment.

2. Equipment

Proper equipment usage, calibration and maintenance represents a similar challenge for biological researchers. Information sources centered on equipment and resulting measurements and their reproducibility are rare. To address this issue, a graduate student in my laboratory began a page on OWW to describe the location and who was in charge of a 96 well microplate reader. In less than a year, the page now contains information on: scheduled user times; creating basic protocols and programs to run experiments; data from control experiments on detection limits, linear range, lamp energy, plate to plate variation, etc; tips such as countering evaporation and arrangement of samples; scripts in Matlab and Excel for data analysis; and a service history of the equipment along with major problems PlateReader. These types of information often cannot be found in other resources. The easy to use interface, along with the power of the wiki software has allowed such informational resources to flourish on OpenWetWare.

3. Biological data aggregation

Frequently, we and other researchers have wanted a quick way to store information gleaned from primary literature. For instance, an OpenWetWare user began populating a page with Escherichia coli genotypes EcoliGenotypes. Later, another user contributed by adding explanations of the cryptic genotype nomenclature allowing those outside the field to more easily understand the information on the page. Such collaborative aggregation of information relevant to a particular topic (in this case E. coli genotypes) is a key feature that OpenWetWare offers in terms of biological databases and informatics. Oftentimes such information aggregates don't initially warrant a structured database but yet are useful. OpenWetWare provides a convenient forum for this data.

Proposal
There are two key priorities that we are addressing in this grant where modest investments in work can increase the quality and scope of the information available on the site. First, while wiki technologies are easy to use, we hope to reduce the learning curve by developing templates and tutorials on building the types of information that users are entering. Second, we hope to build in methods for automatic organization and analysis by developing and refining methods of community categorization, and integrate technologies to incorporate semantic relationships to form more structured, and traversable information.

Goal 1: Templates and tutorials to guide information generation.

Currently OpenWetWare researchers post information primarily from scratch or using existing pages as a reference. In years 1 and 2 of the project, we propose to develop a set of templates and step-by-step tutorials to guide researchers through entering protocols, materials information pages and equipment usage instructions on the wiki. Such help pages will both encourage more complete information pages (by requesting certain pieces of information from users) and catalyze efforts to standardize information organization on the site (making it easier for others to navigate pages).

Goal 2: Extensions for interfacing existing biological databases with OpenWetWare.

Researchers would benefit from being able to link the unstructured information available on OpenWetWare with structured databases such as Entrez, WorldWide Protein Data Bank and organism specific databases like WormBase and FlyBase and others Wheeler-NAR-2003, Berman-NatStrucBiol-2003, Schwarz-NAR-2006, Grumbling-NAR-2006. Initial software extensions (years 1 and 2) would implement automatic link generation on the basis of GenBank accession number or PDB ID much in the way our current Biblio extension works with PubMed (see table X) [need ref to Biblio and Pubmed]. Later in years 3 and 4, these tools will be refined to import data from those databases into OpenWetWare as the user specifies. Hence, OpenWetWare would provide support for researchers interested in mixing structured information sources with the flexibility of OpenWetWare. Since the MediaWiki software package is available under the GPL, any tools we write would be made available to the research community at large. Similarly, we could adopt useful tools from other wiki efforts like EcoliWiki as appropriate Ecoliwiki. This sharing of wiki extensions geared towards the scientific community will be greatly facilitated by the standard wiki distribution described in the second specific aim.

Goal 3: Categorization and tagging of existing pages to enhance searchability of information.

Electronic information is only as useful as our ability to find it when needed. As OpenWetWare grows as an information resource, more extensive metadata information is needed to help users find the information they need. In years 2 and 3 of the project, we will implement software tools that both facilitate tagging pages upon creation and permit categorizing and tagging multiple pages simultaneously. These tags and categories are defined by the users, but allow dynamic organization. For example, if one wants to find all protocols that are related to PCR, the user can execute a search and will find over 100 pages where people discuss the method. However, if users are able to categorize these pages, all the PCR protocols would appear on a general category page automatically linking to the relevant protocols. This dynamic organization will allow OpenWetWare to become a more organized technical resource. We will develop easy ways to categorize and tag particular pages, initially seed a structure for these categorize (that will be modified and extended by the community), and begin the categorization of already existing data.

Goal 4: Semantic web technologies to enhance machine understandability of OpenWetWare content

In years 3-5 of the project, we will move from adding categorization and tagging metadata at the level of individual pages to the use of semantic web technologies to annotate information contained within those pages. Integration of semantic web concepts with wiki's are already being explored as is the utility of the semantic web in biology SemanticMediaWiki. Giving users simple ways to add machine-readable context to information placed within the wiki will provide mechanisms for users to have essentially an easy to use, unstructured, and completely extendable database. We expect that richer annotation of the information in OpenWetWare should lead to a corresponding increase in the utility of the information to the research community.

Overview
OpenWetWare provides researchers with novel tools and opportunities for collaboration. By providing an easy to use and flexible method for developing online communities, the site has already seen the growth of a number of scientific communities, ranging from individual labs to multi-institutional groups working in the same field. Online collaborations overcome many institutional and geographic barriers, removing some of the barriers to collaboration found using traditional approaches. We will provide easier methods for laboratories, and larger biological communities to collaborate on OpenWetWare. In addition, we will provide a software distribution to allow researchers the tools to digitize and organize their own information unsuitable, which will provide easier means to put information online in the future.

Significance
Collaboration in biology is becoming increasingly important as the science and technology driving research has become more complex and is evidenced from the increase in multi-investigator, multi-institutional grants (ref). Many groups contain individuals specializing in computer science, engineering, physics, mathematics, and chemistry to work together on biological problems. These groups have to share biological knowledge, from experimental data to strain locations and history. As biological research scales in size, the traditional methods of information sharing amongst individuals in laboratories become inadequate. Technical resources for collaboration and information sharing between these larger groups usually take the form of internal databases or websites containing necessary information. These implementations often do not take advantage of the latest advances in collaborative information technology, do not scale easily when new types of information need to be shared, and lead to repeated efforts in building out new infrastructure each time a new community is formed.

OpenWetWare has developed tools specifically to foster the growth of new communities online, and this has led an increase from 1 to 60 labs participating in a period of a year. The success of OpenWetWare at fostering user communities cannot be untangled from its success in generating the unique, digitized content described in Specific Aim 1. Results from OpenWetWare user edit analysis shows that users affiliated with labs on OpenWetWare generate XX times the number of edits on average then users that join the site independent of their lab group (JRK: collect stats). We hypothesize that this discrepancy is due to the incentives for contributions implicit in participation in a community. For example, the benefit an individual researcher in isolation receives from placing information on OpenWetWare is mainly in an safe and easy way to store, manage, and organize their own information. This is not very different from an organized individual's laboratory notebook. However, having this information available to a larger community allows that researcher to easily help others, recieve feedback on their own techniques, provide creditable contributions to their labs and larger communities, and easily form collaborations with others. As a consequence of visibility of the information within the commuinty, the quality and quantity of informational resources on OpenWetWare become better.

However, there are some other challenges to providing such information on a site like OpenWetWare. First, creation of a community on OpenWetWare, whether an individual laboratory, or a larger community, initially requires dedicated efforts of a few individuals before a larger set of members join. Second, not all information can be shared openly. Protocols, information, and data that is unpublished often cannot be put onto such sites. For example, within our own laboratory, many individuals run private instances of wikis to store their experimental results. Finally, some communities may need specific tools and extensions that OpenWetWare does not provide in order to make it useful to that community. For example, there are several new proposals for wiki projects for biological information management in particular model organisms such as E. coli and C. elegans.

Here we intend to address these problems in two ways. First, we will develop templates, tutorials, and other mechanisms to make starting community much easier than is currently possible. This will allow communities to join with less work required of the initiating members. Second, we will develop and maintain an open-source distribution of the software and tools that drive OpenWetWare. This will provide users with the ability to privatize or customize their information management needs and will include mechanisms to push back information into the larger community by adding easy mechanisms to publish the information back to OpenWetWare. In addition, this will provide incentives for people to begin putting more information into a standard digitized form, as it will be easier to share this information later as all the individual wikis share the same code.

Preliminary results
In a little more than a year OpenWetWare has grown to include 60 labs as well as a number of multi-institution groups organized around particular areas of research. Different size communities generate different types of information for the site. For instance, individual laboratory communities tend to share practical information such as protocols, details about equipment operations, and results of control experiments. Larger groups share information about community standards, research goals, and shared materials. The examples that follow demonstrate the utility of OpenWetWare for the rapid development of online scientific communities. Furthermore, they help confirm our hypothesis that such communities will be integral in encouraging researchers to contribute novel content to the site.

1. Individual laboratory communities

The most common community on OpenWetWare is the individual research lab, as the usefulness of OpenWetWare as a laboratory information management system (LIMS) is often the initial driver for a laboratory to join. Pam Silver’s group at Harvard Medical School is an excellent example of a lab that has integrated OpenWetWare into their day-to-day research. They use OpenWetWare to share information ranging from lab meeting schedules to details about reagents. As a result of this tight integration, users from the Silver lab alone have contributed more than 30 protocols to OpenWetWare. The tools made available by OpenWetWare enabled the Silver lab to rapidly create an online community site for their lab and to populate it with novel information that otherwise would not have been disseminated to the broader scientific community.

This scenario has been repeated with increasing frequency as more new labs join the site. In particular, work by community members to create a series of tutorial pages geared towards creating user and lab homepages led to a dramatic improvement in the comfort of new users in starting new communities on the wiki. The success in this area has inspired us to create similar tutorial pages in support of the other specific aims, as well as to expand the ease of use of the community development tutorials. Finally, these tutorial pages are also an example of why having a vibrant user community is so essential to the success of OpenWetWare or any project like it – the idea for the tutorial pages was a non-technical solution to the problem of becoming familiar with the wiki. The problem was best addressed by community members who had gone through the process themselves, rather than by a centralized authority, and we will elaborate on how we are more formally harnessing this community leadership in the project organization and logistics section.

2. Subject area communities

Synthetic biology is a relatively new field generating much excitement within biology and engineering. The nascent members of this field have used OpenWetWare as an community organized information portal. The site, http://syntheticbiology.org, is dynamically generated from the community edited wiki page. OpenWetWare is used by researchers in the field for a number of purposes such as: dissemination of recent news relevant to the entire community, discussion pages about experimental protocol standardization, discussions on new research projects and efforts, individual and group research results, links to community resources, conference and job announcements, links to individual labs on OpenWetWare in the field, and much more. (ref) There are five labs that maintain this resource from 4 different institutions. OpenWetWare has provided a common online space for these labs to share information and collaborate more effectively, independent of institutional and geographical barriers. As a result of these labs choosing OpenWetWare to host this resource all information exchanged between groups is freely available to the larger scientific community.

Proposal
The rapid growth of laboratory communities on OpenWetWare has demonstrated the utility of the wiki for enabling the rapid development of online communities. Furthermore, the increased contribution by members participating in these communities suggest that fostering their growth is integral in the success of OpenWetWare. We are proposing two goals to increase the growth of these communities. One is to expand upon our existing tutorials and help pages in order to make the process of starting communities on OpenWetWare even easier. The second is the development of a standard wiki distribution for biological researchers, this distribution will serve groups that would prefer to collaborate off of OpenWetWare whether for technical or philosophical reasons.

Goal 1: Improved tutorial pages for starting new communities Simple tutorial pages already exist for creating lab pages on OpenWetWare, however these could be greatly improved. For instance, taking advantage of a new extension developed by users on the site that allow customization of the viewer for recent changes, limiting it to only pages in your lab group. Additionally, we could develop tutorials for larger communities, for instance how to get your institution on OpenWetWare. Some institutions, such as the University of Chicago, have started doing this on an ad hoc basis, creating pages for some institutional resources, such as an NMR facility. Making this process more clear via tutorials would likely lead to greater adoption of institute communities on OpenWetWare.

Goal 2: OWW-brand Mediawiki distribution

The usefulness of wiki technology has resulted in it being adopted by a number of different scientific groups. Specifically, EcoliHUB and Wormwiki are two recently started community-driven wikis for the E.Coli and Nematode communities. Greater adoption of wikis is an excellent development for scientific communication in general, and OpenWetWare in particular; however it may also result in a fragmentation in the types of wiki software employed by scientists. This could mean that transfer of information between EcoliHUB and OpenWetWare could become difficult, restricting overall opportunities for collaboration. OpenWetWare’s position as the largest scientific wiki (as well as the extensions we have developed specifically for the scientific community, outlined in table X) puts us in a position to spearhead an effort to develop a standard wiki distribution for the scientific community. Leading this effort would have a number of benefits:


 * 1) Preventing the duplication of developer effort in customizing wikis for scientists. By sharing a single distribution, features developed by one community could be shared by all communities.
 * 2) Increasing contribution to OpenWetWare. Since the technology platform is standard, moving information between different OpenWetWare-brand wikis would be seamless.

Many labs on OpenWetWare also maintain private, offline wikis that are restricted to viewing only by lab members or close collaborators. Inclusion of such ‘private pages’ on OpenWetWare would run counter to our mission of enabling free and open sharing of information, however we understand the need for privacy at various stages in the research process. To better support our user communities we will offer the OpenWetWare wiki distribution for labs to run locally. We will also include a “publish to OpenWetWare”-button that would enable seamless uploading of previously private content to the public OpenWetWare pages.

Overview
The same features that make OpenWetWare a powerful tool in research also make it incredibly useful in education. OpenWetWare will have the strongest impact in two key areas of education: laboratory classes and curriculum development. Given OpenWetWare's success so far in courses on the basis of community contributions, dedicated efforts that improve general purpose pages about course hosting along with complementary software tools should further increase OpenWetWare's utility as an educational resource.

Significance
Biology is a constantly changing field that makes it very exciting for researchers, but often makes teaching difficult for educators. Not only are common lab techniques, methods of analysis, and biological details constantly in flux, even the "central dogma" gets turned on its head sometimes (Mattick JS, 2003).

''More background needed here: A case study maybe? modular course development; faster turnover... something!!! please help'' ''Not sure if any of these are what you're looking for but here are a few recent publications describing new ideas for meeting educational goals [], for teaching teachers [] as well as a painfully detailed curriculum development story []. Sounds more like you'd like an article on how the bar is always moving in science, making the teaching of science hard. I can give you a pre-print of an article like that but maybe there are already published ones in JCBE []. '' --NK

There are interesting efforts to integrate openness and web technologies for education. MIT OpenCourseWare is an ambitious effort to digitize all MIT course content for helping self-learners and other educators to develop course content. The are a few problems with this model. First, the rigid structure of course content, hinders rapid course development and reorganization. Second, such an effort requires massive institutional commitments that other universities are either unable or unwilling to do. Third, the process of course development often transcends individual institutional boundaries to gather cutting edge information and research. Finally, for classes such as laboratory classes or advanced seminars there is only an ad-hoc integration with the research community that would keep these classes as cutting edge as they require and give back to the research community.

OpenWetWare, though relatively new, has spontaneously been used by the research community to develop, maintain, and reuse their course curriculum. The ease of development, flexible structure, and reusability indicate that using OpenWetWare and wikis in general for course development can be a very powerful tool. Already, class materials have been updated faster, have much more student input and feedback, have enabled course reuse, and have allowed researchers to donate knowledge and also learn from the classes itself. We intend to encourage further class development, focused on new course development, biological laboratory classes, and advanced seminars in biology by offering ever easier ways to develop course material, taking advantage of the tools developed for previous motivations, and adopt them for educational purposes. If sufficiently promising, we will develop an education-specific distribution that allows others to use these tools independently. (should we include that as an aim then?)

Preliminary results
During semester of 2006, MIT's Biological Engineering department introductory lab techniques class titled Laboratory Fundamentals of Biological Engineering (hereafter referred to by its class number 20.109) was taught using OpenWetWare BE109. This endeavor was a pilot project to assess the utility of using wiki's in general and OpenWetWare in particular to teach undergraduate courses. Given how useful OpenWetWare is in a research lab setting, we expected that it would prove similarly useful in an introductory lab class.

1. OpenWetWare promotes student involvement in class content.

Using a wiki for a laboratory class gives students new avenues for class participation not available to courses that use static websites Wang-2004. Students were given complete edit access over the site. Students weren't merely following a procedure but rather doing experiments and documenting their experiences for their classmates and future years of students BE109gels. From a practical perspective, it improved course materials because students shared the burden of catching and correcting errors in course content BE109errors. Interestingly, some students opted to post their own personal class notes online BE109notes. Thus, OpenWetWare facilitated a two way flow of information: instructors would post information on the wiki and students (at their option) could post their own notes giving us another mechanism to evaluate their understanding. A testament to OpenWetWare's success in engaging undergraduates is that a student from 20.109 now serves as a member on the OpenWetWare steering committee BE109steeringcommittee. Other students continue to use OpenWetWare in their undergraduate research projects BE109urop.

2. OpenWetWare streamlines collaborative teaching and curriculum development.

Like many college courses, 20.109 has a team of people running the course (professors, instructors and teaching assistants). Coordinating among the teaching staff to generate and revise class materials is a very real practical issue. The default solution to this problem is to circulate via email electronic documents allowing one person to edit the document at a time. In OpenWetWare, teachers bypasses this cumbersome process by efficiently sharing and simultaneously editing pages BE109collaborate. While this issue seems trivial, it has a very real impact on the continuity and quality of teaching materials. The easier and more efficiently people can collaborate, the better the resulting work. It also allowed the team to be more responsive to student needs since multiple people could chime in to answer questions BE109questions.

3. Hosting 20.109 on OpenWetWare improves reusability of course materials.

An unexpected but gratifying result of our pilot project of running 20.109 on OpenWetWare is that it made it far easier for others to reuse 20.109 course content in other contexts. Within a month of the end of the semester, there has been interest from outside individuals in taking the section on synthetic biology and transforming it into an independent, standalone unit to introduce people to the emerging field SBstarterkit. Providing course content in a reusable form on OpenWetWare will promote sharing of educational ideas and materials within the community in a way that static course websites cannot. Moreover, having 20.109 protocols and accompanying explanations available on OpenWetWare provide a rich resource for novices entering biological research.

Proposal
Based on the experience of using OpenWetWare in 20.109, it is an enormously valuable tool for education. Its biggest potential impact is on laboratory courses and courses under development. With dedicated effort and software development, we could lower the barriers to hosting courses on OpenWetWare enabling more of the research community to take advantage of this resource for educational purposes.

Goal 1: Develop general templates for common course information.

In years 1 and 2 of the project, significant strides are possible by developing additional help pages regarding courses. For instance as described in aim 1, wiki templates provide a page outline for entering structured information. Such templates have proven their usefulness in other contexts on OpenWetWare by helping new users with page editing and providing substrate for new user and lab pages. Although currently there is no technical barrier to template generation, the nascent teaching community on OpenWetWare lacks the depth needed to implement general purpose page templates for courses. Dedicated resources are needed. Experiences to date suggest that such pages strongly stimulate use of the site. Given this kick-start, the education community will grow, become self-sustaining and maintain these pages via community improvements. Even if such a group fails to take hold, the set of help and template pages will still be a valuable resource to new educators trying to use the site.

Specific deliverables include (a) templates for course syllabi, lecture schedules and homework assignments and (b) explanatory pages for teachers highlighting ways OpenWetWare can be used in courses and (c) a set of help pages for students to help them quickly familiarize themselves with wiki's.

Goal 2: Software tools for courses.

In years 3-5 of the project, we will develop software tools designed to ease transfer of course materials to the site and enhance existing course materials. Most educators compose lectures and handouts in presentation and word processing software like Microsoft Powerpoint and Word. Tools which convert other electronic formats to wiki markup language will ease transitions to OpenWetWare. While some tools like this exist, in practice they fall short of people's needs. Software tools that guide educators through step by step set up of courses on OpenWetWare (wizards) should aid those less at ease with wiki technology. Moreover, the semantic web and categorization technologies proposed in aims 1 and 2 permit enhanced automatic integration of course material with primary research resources on OpenWetWare and elsewhere. Students can then delve deeper into the work at their option. It also provides a path for novice researchers to find course materials. Such integration is key for classes covering newer topic areas in fields like genomics, systems biology and biological engineering in which textbooks are scarce. We would also support efforts to develop an OpenWetWare distribution customized for educational purposes should the demand arise.

Specific deliverables include (a) automatic conversion and upload tools from Powerpoint and Word to wiki format and (b) wizards to guide instructors through course set up on OpenWetWare.

Project organization and logistics
To date OpenWetWare's growth and success is attributable to the volunteer efforts of individual researchers working to make the site a better resource for themselves and the overall community. We have been fortunate to have donations of technical expertise from the MIT Computational and Systems Biology Initiative and monetary support from the MIT iCampus and Microsoft. Under the guidance of the OpenWetWare steering committee, a small team of personnel will be tasked with tool development and maintenance of the site. Tool development will both improve the site's ease of use thereby attracting more of the research community to the site and reduce the workload of the central administrative and technical team. In the long term, we expect relatively little monetary and paid staff support will be necessary to perpetuate the site. Nearly all of the effort in improving the quality of the information available through OpenWetWare will come from the user community. As OpenWetWare grows in importance and scope, we expect that other sources of funding will become available such as donations from private companies.

By making the intellectual property free for reuse and sharing, we hope that OpenWetWare will spark the development of similar scientific information resources leading to both a scientific and technical community that support such endeavors. All software developed under this grant will extend and integrate with the existing MediaWiki software distribution. Therefore, the software that we develop must fall under the terms of the GNU General Public License and thus will be available for free public use and extension. Individual users authorize licensing of the information they contribute to OpenWetWare under terms of the Creative Commons BY-SA license, which allows attributed re-use and extension given that the information is also shared.

Budget justification
Currently, much of the administrative and technical work needed to maintain OpenWetWare is done on a volunteer basis by members of the OpenWetWare steering committee. The steering committee is composed of graduate students, postdoctoral associates, undergraduates and faculty members who have elected to donate their time to ensuring OpenWetWare's success because they believe it represents an important biological information resource and tool. However, as OpenWetWare grows adminstrative and technical responsibilities will need to be transferred to a small group of paid staff. This transfer will have several advantages. First, it will allow stabilization and continuity in the backend software and technical support for OpenWetWare assuring its reliability. Second, it will free the steering committee to focus their efforts on improving the quality of the information on OpenWetWare and its utility to the research and education communities. Third, it will enable pursuit of development projects that will enhance the richness of information on OpenWetWare while retaining its flexibility, ease of use and convenience that the research community needs.

We are requesting funds for a part-time administrative assistant for years 1 and 2 whose position should mature into a full-time project lead for years 3-5. This person will facilitate coordination of software development priorities and implementation between the steering committee and the development team, field email correspondence from site users and administer the grant. A substantial travel grant is included to enable this person to travel to various labs and institutions to educate researchers about the availability and usefulness of the resource while also gathering feedback from OpenWetWare users on their experiences with the site. Include a sentence about how much personnel like this cost.

Funds are also requested for two full-time software development personnel for years 1-5. These programmers will be responsible for developing tools that makes it easier for researchers to generate, organize and annotate content on OpenWetWare. The ease of use of the site is crucial to the quality and quantity of information on OpenWetWare available to the research community. Include a sentence about how much personnel like this cost.

My lab and others on OpenWetWare already depend critically on OpenWetWare's reliability and uptime for day to day research needs. As OpenWetWare grows, the same will be true for researchers around the world. Thus, this proposal requests funds to outsource system administration to TechSquare, Inc. TechSquare, Inc. currently maintains the servers that run OpenWetWare via donated financial support from MIT's Computational and Systems Biology Initiative. Dedicated funds for OpenWetWare's server maintenance and support will ensure regular backups, timely system updates and low downtimes in the longterm.

Finally, we request funds for hardware to ensure that OpenWetWare can continue to meet worldwide demand for site edits and views. Funds will be needed in year 1 to upgrade existing hardware to meet OpenWetWare's current growth and again in year 3 to meet the expected growth from the proposed work.

Placeholder table for expenditure amounts
Need to get numbers everywhere there is an X

=Facilities, equipment and other resources=

Major equipment
OpenWetWare is currently hosted on a Dell 1850 server with Dual 3.2 GHz Intel Xeon, 800 Mhz FSB and 2GB memory professionally hosted by Techsquare, Inc, and physically located in the CSBi server room. In addition, the Endy group has two additional servers. The first is a Dell Precision 530 Dual 2 GHz Intel Xeon with 1GB memory. The second is a dual Pentium III 600Mhz with 500MB of RAM which runs a redundant copy of OpenWetWare for development purposes. Both servers are connected to the internet through the CSBi Reseach Network.

Other resources
Techsquare, Inc has supported the CSBi Computing infrastructure for three years and provides a valuable systems administration role. Techsquare has worked with many departments at MIT and have extensive expertise with the MIT computing environment.