BAMS Principles and Policies (version 1.0)

Fundamental Principles

The grounding principles of the Brain Architecture Management System are openness to the neuroinformatics and neuroscience communities, and collaboration with other neuroinformatics and neuroscience research groups. These two principles are implemented in all components of BAMS: database structure, collation methodology of data and metadata, user interfaces, and inference engines. All these components are documented in our published papers, Online Manual, and additional help pages. All collated data and metadata are available to the public in different electronic forms.

BAMS is information and knowledge provider to more than 15 research groups including NIF, and many of these have direct and full access to its backend database structure.

We seek to make available the structure and contents of BAMS in the newest standards available that allow dissemination and integration of knowledge (e.g. Semantic web, RDF/OWL 2.0). We actively participate to groups that work toward constructing integrated vocabularies, and standards in (e.g. PONS).

Data and metadata collation

Our guiding principles for data and metadata collation are precision and openness. Since BAMS' main role is to archive and understand neural circuitry, the precision of brain parts definitions and their hierarchies and of neuroanatomical mapping of experimental results are crucial in our collation activity.

We seek to insert high quality information from primary published literature. We faithfully associate citations and PubMed links to each collated datum and strive to keep the collated data as they were reported in the associated references. We insert collators' interpretations of collated data only when the published reports are imprecise in terms of neuroanatomical mapping (see Bota & Swanson, 2010 for details). These interpretations are always explained with additional metadata, as well as in full text.

We welcome participation of neuroscientists to our collation effort, and we believe that population of BAMS can be accomplished only through a collective effort. We have made publicly available Excel forms for collation of neuroanatomical projections from the literature, and the Personal Workspace of BAMS can be used to insert the presence and absence of molecules in the rat central nervous system. Our standards and methods for collation of data and metadata are documented in Bota et al. 2005, Bota & Swanson, 2006, Bota & Swanson, 2007, Bota & Swanson, 2008, and Bota & Swanson, 2010.

Accessibility of structure and data

The structure of BAMS is fully explained in our published papers, and in the Online Manual. The structure of BAMS is also available to different research groups, via backend database connections. Finally, we have started to make available BAMS' structure in RDF/OWL format, which is the backbone of the Semantic Web.

All inserted data and metadata are available to the public, either through BAMS' interface that includes multiple search forms, or through BAMS' inference engines. Data and metadata are also available in Excel and XML formats, as well in the Semantic Web standard RDF/OWL format.

Conditions of use and limitations

We emphasize the collaboration between neuroinformatics and neuroscience groups, and we are open to requests for sharing BAMS' structure and collated data. Although we prefer the use of OWL standards provided by BAMS to populate new systems, we are fully aware that different systems may be constructed in the future, for different needs.

Please contact us before using any part of BAMS or data therein, for constructing and populating new systems.

Because each part of BAMS is documented in published references, please cite the appropriate ones whenever any part of BAMS is used for development of a different system, or data provided by BAMS is used for populating other systems, or for analysis.

We are fully responsible for the data and knowledge (structured information) inserted in BAMS. However, we cannot guarantee its integrity when inserted in other systems, unless our names are associated with each imported datum.