KBB: A Knowledge-Bundle Builder for Bio-Research

We propose research into and development of a “Knowledge-Bundle Builder for Bio-Research.” We direct our proposed research at the broad Challenge Area 04: Clinical Research and the specific Challenge Topic 04-NS-102 Developing web-based entry and data-management tools for clinical research. The volume of biological data is enormous and increasing rapidly. Unfortunately, the information a bio-researcher needs is scattered in various repositories and in the published literature. To do activities bio-researchers need a system that can efficiently locate, extract, and organize available bio-information so that it can be analyzed and scientific hypotheses can be verified. Currently, bio-researchers manually search for information of interest from thousands of data sources (either online repositories or publications) to achieve their goals. This process is tedious and time-consuming. As a specific example, to do a recent study about associations between lung cancer and TP53 polymorphism, researchers needed to: (1) do a keyword-based search on the SNP data repository for “tp53” within organism ”homo sapiens”; (2) from the returned records, open each record page one by one and find those coding SNPs that have a minor allele frequency greater than 1%; (3) for each qualifying SNP, record the SNP ID and many properties of the SNP; (4) perform a keyword search in PubMed and skim the hundreds of manuscripts found to determine which manuscripts are related to the SNPs of interest and fit their search criteria;1 and (5) extract the information of interest (e.g., the statistical information, patient information, and treatment information) and organize it. In an effort to automate some of this manual tedium and speed up the search and extraction process, bio-information-retrieval researchers have worked on finding relevant documents (e.g., [BFS06, CJR+07]), but this work is directed only to specific topics (e.g., both [BFS06] and [CJR+07] focus on locating SNPs only). The research challenge for doing high-precision document filtering even for specific topics is huge, and it is even more of a challenge to generalize these ideas. How can a system do high-precision document filtering for any bio-research topic? Further, when found, how can we mitigate the tedium of extracting and organizing the relevant information so as to facilitate analysis and decision making? The key is to answer both of these questions in unison—extract to identify and identify to extract. We take on this research challenge and propose here the idea of a Knowledge Bundle (KB) and a Knowledge-Bundle Builder (KBB). As we explain below, a KB includes an extraction ontology, which allows it to both identify and extract information with respect to a custom-designed schema. Construction of a KB itself can be a huge task—but one that is mitigated by the KBB. Construction of the KB under the direction of the KBB proceeds as a natural progression of the work a bio-researcher does in manually identifying and gathering information of interest. As a bioresearcher begins to work, the KBB immediately begins to synergistically assist the bio-researcher and quickly “learns” and is able to take over most of the tedious work. As mentioned in the first paragraph, our proposed KB and KBB address the challenge area of clinical research and specifically address the topic challenge 04-NS-102. We quote from 04-NS-102 in terms of our research agenda: We propose “developing web-based entry and data-management tools for clinical research. The [tools] are to be open source [and to provide for] user-friendly, webbased data entry and data management. [They can] be customized by investigators [and could] serve as a core resource for the community.” Further, the research we propose has the potential