Bioinformatics is quickly becoming the central core
that integrates the many disparate bodies of data, scientific knowledge and computational infrastructure from fields as diverse as genetics, structural biology, medical and animal models of disease, imaging, engineering, etc. For example, biological imaging and shape representation 
, sequence analysis 
, regulatory genomics 
and alternative splicing 
are critical components of computational biology. Many space, time, function or interaction modeling techniques apply across the vast spectrum of scales from genotypes to phenotypes, from the small scale of microarray imaging for genomics, to the larger in vivo
In this manuscript, we discuss challenges of development, maintenance and dissemination of robust computational biology resources (data, tools and services, see below). The rapid growth, usage specifications, application domain and method diversity of contemporary computational biology resources presents an impediment to finding, comparing and integrating such data and tools. We propose a new decentralized, extensible, lightweight, scalable and portable framework (iTools
) to enable resource location, management, evaluation and mediation. iTools
aims to provide an open, yet secure, infrastructure for resource archival, retrieval and traversal; communication between tool developers, clinicians, scientists and the general community; and to facilitate the interoperability of resources developed by a wide spectrum of investigators and employed in a variety of computational biology studies. There is an increasing need in the bioinformatics community for accurate accounting of data and resources, e.g., the bioinformatics Resourceome effort 
We leveraged the broad research, computational, applied and scientific expertise available at the seven National Centers for Biomedical Computing (NCBC) to design, implement and validate a new schema for classifying, categorizing and integrating different computational biology resources across space-and-time scales, biomedical problems, computational infrastructures and mathematical foundations. The proposed infrastructure is not limited to needs and applications of the NCBCs. Rather it is intentionally designed to be extensible and applicable to broader range of biomedical research.
National Centers for Biomedical Computing (NCBCs)
Seven NCBCs were funded by NIH as part of the Roadmap for Medical research in 2004 and 2005. The computational biology research at these centers and their partner and collaborating institutions spans a wide range, which ensures that the NCBC program provides a foundation for a National infrastructure for diverse biomedical computing. The research and clinical goals, project descriptions, home-pages and the computational infrastructures provided by of all NCBC centers is available online at http://www.NCBCs.org
Current approaches for software tool classification, categorization and integration
There are a number of efforts to develop integrated resources that provide a common user-friendly access to variety of tools and software programs for computational biology. summarizes many of these attempts. These are important efforts to catalog computational tools; however, there are remaining challenges, whose solutions may significantly affect both the development of new tools and technologies and the utilization of existent novel computational biology resources. For example, there is no one-stop site for researchers to go to search, contrast and integrate tools. Many sites specialize on certain types of tools, and researchers must know about these sites to find the tools they seek. In addition, graphical traversal, semantic search, and effective interfaces for tool integration are still mostly envisioned for the future. Research progress also may be significantly potentiated by providing agile interfaces, extensibility and portability features to any new meta-resource of computational biology tools and software.
Examples of Meta-Resources for Computational Biology.
Goals of this research
Our overarching aim is to build a meta-resource that catalogs computational biology data, tools and services provided by the computational biology community. More specifically, our goal is to implement iTools as an extensible meta-resource, validate it using NCBC resources, and enable its growth in depth and breadth in the future. Additional goals of the iTools resource are to:
Develop a general, scalable and decentralized ontology
of computational biology resources (NCBC Biomedical Resource Ontology http://Bioportal.Bioontology.org
), which can be used in the context of a flexible XML resource description and hastens the creation of plug-ins and an interface infrastructure for utilization by an external computer program.
Develop a database
of XML descriptions of computational biology resources (adhering to an XML data type) general enough to describe most types of existing resources. This database should allow an easy and robust access (manual or computer-initiated) to the resource descriptions, include powerful and user-friendly search and download interface, provide an agile syntax of the XML descriptors, and facilitate the meta-resource curation, community editing and evaluation. The latest version (V4) of this XML is located at the NCBC biositemap site www.biositemap.org
Develop the infrastructure for integration and interoperability of meta-algorithms for biomedical computing. This is a high-level goal that aims to facilitate the communication between different tools in terms of meta-data resource descriptions (e.g., software tool I/O types) useful for streaming data from one resource to the next (perhaps facilitated by data mediation modules).