Much as a bound laboratory notebook (see Figure 1 below) allows researchers to keep track of relevant data in the regular execution of their work, an ELN provides low-level computational support for all the information that researchers might place into their notebooks. This may include important observations, conclusions and interpretations as well as details that are vital to the execution of the scientific work but would be considered irrelevant to the final ‘punchline’ of the study (such as purchase orders for reagents, or other logistical items). This fits in with our perspective of science being a process of inscription, either into a notebook or a computer program. ELN are useful for the simple fact that data must, at some point, be entered into a computer so that it is then retrievable. When the original researcher enters data as a natural part of the scientific process, then populating the target system is straightforward. This holds true for molecular biology, since one of the prerequisites for publishing a report of gene or protein is that the relevant data must be uploaded to one of the large scale consortia databases. If, as is the case for most centralized neuroinformatics systems, the data are collated by a single individual or team from non-digital sources, then database population becomes very challenging. The widespread use of reliable ELN software would greatly ease this process, and this is underscored by the recent data sharing policy announcement of the U.S. National Institutes of Health (NIH). This policy specifies that all data generated with federal funding exceeding $500K in any single year must be made publicly available (http://grants.nih.gov/grants/policy/data_sharing/).

Figure 1. A typical example of a laboratory notebook showing notes taken by a scientist.

There exist many robust commercial ELN applications (e.g., Amphora: http://www.amphora-research.com, Axiope: http://www.axiope.com, etc.) as well as open-source applications within academia, such as the Environmental Molecular Sciences Laboratory’s Electronic Laboratory Notebook (http://collaboratory.emsl.pnl.gov/), and the NeuroSys system (http://neurosys.cns.montana.edu/). These systems have a common design: a local server acts as a repository for data files and structured annotations so that they may be added, queried and manipulated, usually over the web. In effect, these systems provide an asset management service, by offering a local store for all local laboratory-based information. Key design issues for these systems are flexibility, ease of use, robustness and scalability to ensure their usability within the working environment of the laboratory. The NeuroScholar system also contains an ELN component. Common technology of interest to these systems are standardized libraries for file transfer and storage, such as Jakarta’s Slide project based on Web-based Distributed Authoring and Versioning (WebDAV) methods and the San Diego Supercomputing Center’s Storage Resource Broker (SRB). These technologies operate as middleware that permit users to collaboratively edit, manage and annotate files on remote web servers.

Future Development
We aim to extend Neuroscholar with NLP methods to make the process of data entry as seamless as possible enabling rapid development of an ELN prototype.

Demos
The ELN demo can be accessed here.

Software Downloads
The ELN is available for download from here.

Documentation

• ELN usage guide 

Publications