J. Steven Hughes

Bioinformatics: Biomarkers of early detection

Capturing, sharing, and publishing cancer biomarker research data are all fundamental challenges of enabling new opportunities to research and understand scientific data. Informatics experts from the National Cancer Institutes (NCI) Early Detection Research Network (EDRN) have pioneered a principled informatics infrastructure to capture and disseminate data from biomarker validation studies, in effect, providing a national-scale, real-world successful example of how to address these challenges. EDRN is a distributed, collaborative network and it requires its infrastructure to support research across cancer research institutions and across their individual laboratories. The EDRN informatics infrastructure is also referred to as the EDRN Knowledge Environment, or EKE. EKE connects information about biomarkers, studies, specimens and resulting scientific data, allowing users to search, download and compare each of these disparate sources of cancer research information. EKEs data is enriched by providing annotations that describe the research results (biomarkers, protocols, studies) and that link the research results to the captured information within EDRN (raw instrument datasets, specimens, etc.). In addition EKE provides external links to public resources related to the research results and captured data. EKE has leveraged and reused data management software technologies originally developed for planetary and earth science research results and has infused those capabilities into biomarker research. This paper will describe the EDRN Knowledge Environment, its deployment to the EDRN enterprise, and how a number of these challenges have been addressed through the capture and curation of biomarker data results.

Enabling effective curation of cancer biomarker research data

The dramatic increase in data in the area of cancer research has elevated the importance of effectively managing the quality and consistency of research results from multiple providers. The U.S. National Cancer Institutes Early Detection Research Network (EDRN) is a prime example of a virtual organization, sponsoring distributed, collaborative work at dozens of institutions around the country. As part of a comprehensive informatics infrastructure, The NASA Jet Propulsion Laboratory, in collaboration with Dartmouth Medical School, has developed a web application for the curation of cancer biomarker research results. In this paper, we describe and evaluate the application in the context of the EDRN content management process, and detail our experience using the tool in an operational environment to capture and annotate biomarker research data generated by the EDRN.

The Planetary Data System. A Case Study in the Development and Management of Meta-Data for a Scientific Digital Library

The Planetary Data System (PDS) is an active science data archive managed by scientists for NASAs planetary science community. With the advent of the World Wide Web, the majority of the archive has been placed on-line as a science digital library for access by scientists, the educational community, and the general public. The meta-data in this archive, primarily collected to ensure that future scientists would be able to understand the context within which the science data was collected and archived, has enabled the development of sophisticated on-line interfaces. The success of this effort is primarily due to the development of a standards architecture based on a formal model of the planetary science domain. A peer review process for validating the meta-data and the science data has been critical in maintaining a consistent archive. In support of new digital library research initiatives, the PDS functions as a case study in the development and management of meta-data for science digital libraries.

Ontology-based information model development for science information reuse and integration

Scientific digital libraries serve complex and evolving research communities. Justifications for the development of scientific digital libraries include the desire to preserve science data and the promises of information interconnectedness, correlative science, and system interoperability. Shared ontologies are fundamental to fulfilling these promises. We present a tool framework, a set of principles, and a real world case study where shared ontologies are used to develop and manage science information models and subsequently guide the implementation of scientific digital libraries. The tool framework, based on an ontology modeling tool, has been used to formalize legacy information models as well as design new models. Within this framework, the information model remains relevant within changing domains and thereby promotes the interoperability, interconnectedness, and correlation desired by scientists.

Architecting Data-Intensive Software Systems

Data-intensive software is increasingly prominent in today’s world, where the collection, processing, and dissemination of ever-larger volumes of data has become a driving force behind innovation in the early twenty-first century. The trend towards massive data manipulation is broad-based, and case studies can be examined in domains from politics, to intelligence gathering, to scientific and medical research. The scientific domain in particular provides a rich array of case studies that offer ready insight into many of the modern software engineering, and software architecture challenges associated with data-intensive systems.

An Ontology-Based Archive Information Model for the Planetary Science Community

The Planetary Data System (PDS) information model is a mature but complex model that has been used to capture over 30 years of planetary science data for the PDS archive. As the de-facto information model for the planetary science data archive, it is being adopted by the International Planetary Data Alliance (IPDA) as their archive data standard. However, after seventeen years of evolutionary change the model needs refinement. First a formal specification is needed to explicitly capture the model in a commonly accepted data engineering notation. Second, the core and essential elements of the model need to be identified to help simplify the overall archive process. A team of PDS technical staff members have captured the PDS information model in an ontology modeling tool. Using the resulting knowledge-base, work continues to identify the core elements, identify problems and issues, and then test proposed modifications to the model. The final deliverables of this work will include specifications for the next generation PDS information model and the initial set of IPDA archive data standards. Having the information model captured in an ontology modeling tool also makes the model suitable for use by Semantic Web applications.

Policy Recommendations for Institutions to Serve as Trustworthy Stewards of Research Data

The Open Archival Information System (OAIS ) Reference Model, published as ISO 14721, has been adopted as the “de facto” standard for systems that preserve data. ISO 16363, the standard for Audit And Certification Of Trustworthy Digital Repositories , is based on ISO 14721 and contains the criteria for auditing various kinds of repositories in terms of their potential to provide trustworthy services for data management and preservation . Institutions that manage repositories for research data need to attain compliance with ISO 16363 if they plan to serve as trustworthy digital repositories . As an initial step, institutions that operate repositories for managing and preserving research data should create and follow policies to address the ISO 16363 requirements.Recommendations are offered for establishing and implementing policies within institutions that plan to serve as trustworthy repositories of research data holdings .

Developing an open source, reusable platform for distributed collaborative information management in the Early Detection Research Network

For the past decade, the NASA Jet Propulsion Laboratory, in collaboration with Dartmouth University has served as the center for informatics for the Early Detection Research Network (EDRN). The EDRN is a multi-institution research effort funded by the U.S. National Cancer Institute (NCI) and tasked with identifying and validating biomarkers for the early detection of cancer. As the distributed network has grown, increasingly formal processes have been developed for the acquisition, curation, storage, and dissemination of heterogeneous research information assets, and an informatics infrastructure has emerged. In this paper we discuss the evolution of EDRN informatics, its success as a mechanism for distributed information integration, and the potential sustainability and reuse benefits of emerging efforts to make the platform components themselves open source. We describe our experience transitioning a large closed-source software system to a community-driven, open source project at the Apache Software Foundation, and point to lessons learned that will guide our present efforts to promote the reuse of the EDRN informatics infrastructure by a broader community.

A Reference Architecture for Space Information Management

We describe a reference architecture for space information management systems that elegantly overcomes the rigid design of common information systems in many domains. The reference architecture consists of a set of flexible, reusable, independent models and software components that function in unison, but remain separately managed entities. The main guiding principle of the reference architecture is to separate the various models of information (e.g., data, metadata, etc.) from implemented system code, allowing each to evolve independently. System modularity, systems interoperability, and dynamic evolution of information system components are the primary benefits of the design of the architecture. The architecture requires the use of information models that are substantially more advanced than those used by the vast majority of information systems. These models are more expressive and can be more easily modularized, distributed and maintained than simpler models e.g., configuration files and data dictionaries. Our current work focuses on formalizing the architecture within a CCSDS Green Book and evaluating the architecture within the context of the C3I initiative.