Essie: A Concept-based Search Engine for Structured Biomedical Text

Abstract

This article describes the algorithms implemented in the Essie search engine that is currently serving several Web sites at the National Library of Medicine. Essie is a phrase-based search engine with term and concept query expansion and probabilistic relevancy ranking. Essie's design is motivated by an observation that query terms are often conceptually related to terms in a document, without actually occurring in the document text. Essie's performance was evaluated using data and standard evaluation methods from the 2003 and 2006 Text REtrieval Conference (TREC) Genomics track. Essie was the best-performing search engine in the 2003 TREC Genomics track and achieved results comparable to those of the highest-ranking systems on the 2006 TREC Genomics track task. Essie shows that a judicious combination of exploiting document structure, phrase searching, and concept based query expansion is a useful approach for information retrieval in the biomedical domain.

NCBO Resource Index: Ontology-Based Search and Mining of Biomedical Resources

“…HealthCyberMap [31] uses ontologies and semantic distances for visualizing biomedical resources information. Essie [32] shows that a judicious combination of exploiting document structure, phrase searching, and concept based query expansion is useful for domain optimized information retrieval. Finally, other studies such as Khelif and colleagues [33] illustrate the annotation of a specific resource with specific ontologies (the GeneRIF resource annotated with UMLS and Galen in this case).…”

Tasks, topics and relevance judging for the TREC Genomics Track: five years of experience evaluating biomedical text information retrieval systems

“…The TREC Genomics track differs from other biomedical challenges in that the main focus of the ad hoc task was to develop systems based on real needs of biomedical researchers, as opposed to the needs of database curators (classification tasks in the track did address curation needs, but they will not be discussed here; see Cohen and Hersh 2006). Several publications by TREC participants discuss features of their systems, such as query construction, differences in indexing, and use of ontologies (Ide et al 2007). Instead, the aim of this work is to describe how the annotation used to evaluate the systems was created, and how each year's experiences drove the evolution of corpora, tasks and annotation methods.…”

Infectious Disease Ontology

“…The greatest success has come from the assignment of terms from vocabulary resources to individual documents within large collections, a process referred to as indexing. MeSH has long been used to index documents within the PubMed database [26], and, more recently, ontologies have been used for this purpose, allowing text-10 mining algorithms to take advantage of the richer set of relations and their formal definitions [27,28,29]. The identification of documents that are relevant to a query within a collection (document retrieval) is greatly facilitated by utilizing the ontologies' structure.…”

“…Genome-level characters: Features of a genome or its products other than the linear sequences of nucleotides or amino acids that can be assessed for phylogenetic analysis 27 Genomics: The study of the entire genome of an organisms; structural genomics includes whole-genome sequencing, whereas functional genomics aims to determine the functions of all genes Genotype: The entire genetic constitution of an organism or the genetic composition at a specific gene locus or set of loci Grid: A fully distributed, dynamically reconfigurable, scalable and autonomous inftrastructure to provide location independent, secure and efficient access to a coordinated set of services encapsulating and virtualizing resources Informed consent: A legal term referring to a situation where a person can be said to have given his or her consent based upon an appreciation and understanding of the facts and implications of an action Health Level 7 (HL7): A health data interchange standard designed to facilitate the transfer of health data resident on different and disparate computer systems in a health care setting Homoplasy: A pattern of character states that supports an alternative to the true, accepted or most parsimonious evolutionary tree that is generally caused by evolutionary changes Horizontal gene transfer: Any process in which an organism transfers genetic material to another cell that is not its offspring. This process is in contrast to more common vertical gene transfer, which occurs when genetic information is passed from parent to offspring Infectome: System of networks of interacting host and pathogen's genes, proteins and metabolites involved in a process of infection and disease Intron: Portions of a gene between the coding exons that are also transcribed, but are enzymatically removed from the mRNA before its translation into a protein Knowledge base: A repository for the knowledge used by a knowledge system Knowledge based system: A computer system that represents and uses knowledge to carry out a task…”