Smith Professor Wins $2M NSF Grant for Research in Data Management for the Life Sciences

Dr. Louiqa Raschid, professor of information systems at the University of Maryland's Robert H. Smith School of Business, is part of a team of researchers awarded a $2-million grant from the National Science Foundation (NSF) for data management research for the life sciences. Rashid's research objective is to develop new data modeling and data integration tools that will make scientists and biologists more effective and efficient as they navigate through vast amounts of Web-accessible information.

Rashid's research will focus on reducing the manual steps and labor intensiveness associated with scientific research through the development of tools that capture a scientists domain knowledge and use this knowledge to suggest useful queries and to rank results. These tools will enable scientists to quickly identify information closely tied to their research, versus information that is irrelevant, says Raschid.

We all depend on Google to find useful information, says Raschid. While Google does index a significant portion of the Web, it does not index specialized data sources that have knowledge on genes and proteins. Scientists also need to combine information in a meaningful way from multiple data sources and there are few tools to support data integration in scientific disciplines.

The NSF grant is for a period of three years (2004-2007), and is an example of the ongoing effort to support data management and to improve the completeness and quality of data within the scientific community (SEIII Program). Life science data sources are inherently heterogeneous and support multiple data types including full text, images, semi-structured data, relational data, and flat files. They cannot be effectively integrated using standard DBMS technology, since this technology was developed to primarily support business data, says Raschid.

I will apply data modeling, data management and data analysis to the highly interconnected data entries and links, to identify data that is most relevant to a particular research question, says Raschid. It is important to ensure that any automated platform for data integration supports the scientist, but does not attempt to substitute for the domain knowledge of the scientist, or otherwise make choices or control decisions for the scientist, she adds.

To develop the data modeling and integration tools described, Raschid works closely with life sciences researchers, like those at the Memorial Sloan Kettering Cancer Center and the National Institute of Health's National Center for Biotechnology Information (NCBI). For example, there is a current project to extract all the links between entries in PubMed (the citation index for all medical literature) and the Human Genome, to analyze these links, and to label them with domain specific semantics. Raschid uses data modeling and data management techniques to view trends and patterns in data, and to generate effective and efficient navigation paths for scientists to integrate data across Web-accessible sources.

Raschid has extensive experience, academically and professionally, in the data modeling, data management, and integration fields. She received a Ph.D. in electrical engineering from the University of Florida, Gainesville, in 1987. At the Smith School, she holds a joint appointment with the Institute for Advanced Computer Studies, the Department of Computer Science, and the Center for Bioinformatics and Computational Biology. She has been recognized by the vice president of research at the University of Maryland to be among the Top 100 Faculty in obtaining external research funding. Raschid serves on the editorial board of ACM Computing Surveys, the IEEE Transactions on Knowledge and Data Engineering, the Journal of Very Large Data Bases, and the INFORMS Journal of Computing. She is a member of IEEE, ACM, and the Society of Women Engineers.

Other faculty on the team include Lise Getoor, assistant professor of computer science, University of Maryland; Zoe Lacroix, Arizona State University; and Terry Gaasterland, University of California San Diego and Rockefeller University.

Smith Media Group, Monisha Banerjee, MBA Candidate 2006, contributed to this article.