Presentation about Text Mining from Biomedical Literature and Conceptual Biology
p53 is expressed as a central Concept connected to broader, narrower, synonymous and co-occurring Concepts as defined by the National Library of Medicine’s (NLM) Unified Medical Language System© (UMLS) Metathesaurus©. Developing Concepts from UMLS ensures that our application "understands" the meaning of the language of biomedicine and health.
CREB-Binding Protein is expressed as a central Concept connected to co-occurring Concepts mined from the over 16 million articles in the MEDLINE© database dating from 1950 to 2003. The co-occurring Concepts and their semantic type are stored in the Araicom Knowledge Warehouse.
Text mining is the science of processing and analyzing text collections to discover potentially useful and interesting knowledge. Araicom excels at quickly mining hidden connections among biomedical Concepts from large amounts of scientific literature.
Current research on conceptual biology focuses on hypothesis generation from biomedical literature. To implement the potential of Conceptual Biology, Araicom developed a conceptual biology research platform, which supports generating and conceptually testing multiple types of biomedical hypotheses.
Organizing traditionally published literature into digital libraries and providing efficient indexing and retrieval tools over these libraries has had the consequence of making an unprecedented amount of information and knowledge readily available. To harvest this knowledge, we must find ways to leverage our technological awareness and learning. This need has recently opened the door for human assistance requirements through machine intelligence. Knowledge extraction, text mining and conceptual relationships are key parts of the sciences involved in processing and analyzing text collections to discover potentially useful and interesting knowledge. Discovering new knowledge from scientific literature using these and other emerging text mining methods is termed as literature-based discovery.
The foundational paper co-authored by Araicom and its academic collaborators. The primary focus is the presentation of results from the Conceptual Biology and Literature Discovery research collaborative project. The paper provides background for the relevant sciences and challenges in prior work, innovations and validations with examples of the current work, and considerations for future research directions.
This is the acknowledged source of the term "Conceptual Biology" as applied to the use of computational methodologies in sufficiently large data collections. This essay is credited with pointing out clearly how basic computational services can be useful in producing new approaches to the comprehension of large scale repositories. The essay explores computational examples related to p53, a significant protein related to cancer research, and points to the emergence of computational investigation of the literature as an appropriate and desirable basis for research based on the empirical evidence of the peer-reviewed literatures.
This paper is a recent review of the field highlighting the current status of research in this area. It includes a discussion of the strengths, weaknesses, and challenges attributed to the emerging field of Conceptual Biology. Much of the prior work in Conceptual Biology is highlighted, and an overview discussion of the various academic approaches to the architecture and deployment of the Pairwise Algorithm in biomedical literature research is reviewed.
This paper presents one of the variations of the Pairwise algorithm implementations. The authors speak as Subject Matter Experts on the issues involving the details of the biomedical literatures and the UMLS® and MeSH® knowledge systems. A key perspective provided by this paper is found in the narrative and references to the prior works in Conceptual Biology; the issue of clinical validation of the computational research is very well documented.
Another well executed paper with clear discussion and presentation of approach and results from the research team’s application of the Pairwise algorithm. Of significant interest is the hypotheses targeting based on thalidomide. This paper was instrumental in pointing to the potential therapeutic applications of thalidomide as an anti-inflammatory. Thalidomide research is now at the center of a still growing $2.5 billion biomedical research and development space with three major biotech companies involved.