Biomedical subjects

W John Wilbur

Publications and source records attributed to W John Wilbur.

12 recordsLinked to original sources

PubMed2006

New directions in biomedical text annotation: definitions, guidelines and corpus construction.

BACKGROUND: While biomedical text mining is emerging as an important research area, practical results have proven difficult to achieve. We believe that an important first step towards more accurate text-mining lies in the ability to identify and characterize text that satisfies various types of information needs. We report here the results of our inquiry into properties of scientific text that have sufficient generality to transcend the confines of a narrow subject area, while supporting practical mining of text for factual information. Our ultimate goal is to annotate a significant corpus of biomedical text and train machine learning methods to automatically categorize such text along certain dimensions that we have defined. RESULTS: We have identified five qualitative dimensions that we believe characterize a broad range of scientific sentences, and are therefore useful for supporting a general approach to text-mining: focus, polarity, certainty, evidence, and directionality. We define these dimensions and describe the guidelines we have developed for annotating text with regard to them. To examine the effectiveness of the guidelines, twelve annotators independently annotated the same set of 101 sentences that were randomly selected from current biomedical periodicals. Analysis of these annotations shows 70-80% inter-annotator agreement, suggesting that our guidelines indeed present a well-defined, executable and reproducible task. CONCLUSION: We present our guidelines defining a text annotation task, along with annotation results from multiple independently produced annotations, demonstrating the feasibility of the task. The annotation of a very large corpus of documents along these guidelines is currently ongoing. These annotations form the basis for the categorization of text along multiple dimensions, to support viable text mining for experimental results, methodology statements, and other forms of information. We are currently developing machine learning methods, to be trained and tested on the annotated corpus, that would allow for the automatic categorization of biomedical text along the general dimensions that we have presented. The guidelines in full detail, along with annotated examples, are publicly available.

Computational Biology↗

PubMed2006

Using MEDLINE as a knowledge source for disambiguating abbreviations and acronyms in full-text biomedical journal articles.

Biomedical abbreviations and acronyms are widely used in biomedical literature. Since many of them represent important content in biomedical literature, information retrieval and extraction benefits from identifying the meanings of those terms. On the other hand, many abbreviations and acronyms are ambiguous, it would be important to map them to their full forms, which ultimately represent the meanings of the abbreviations. In this study, we present a semi-supervised method that applies MEDLINE as a knowledge source for disambiguating abbreviations and acronyms in full-text biomedical journal articles. We first automatically generated from the MEDLINE abstracts a dictionary of abbreviation-full pairs based on a rule-based system that maps abbreviations to full forms when full forms are defined in the abstracts. We then trained on the MEDLINE abstracts and predicted the full forms of abbreviations in full-text journal articles by applying supervised machine-learning algorithms in a semi-supervised fashion. We report up to 92% prediction precision and up to 91% coverage.

Artificial Intelligence↗

PubMed2005

GENETAG: a tagged corpus for gene/protein named entity recognition.

BACKGROUND: Named entity recognition (NER) is an important first step for text mining the biomedical literature. Evaluating the performance of biomedical NER systems is impossible without a standardized test corpus. The annotation of such a corpus for gene/protein name NER is a difficult process due to the complexity of gene/protein names. We describe the construction and annotation of GENETAG, a corpus of 20K MEDLINE sentences for gene/protein NER. 15K GENETAG sentences were used for the BioCreAtIvE Task 1A Competition. RESULTS: To ensure heterogeneity of the corpus, MEDLINE sentences were first scored for term similarity to documents with known gene names, and 10K high- and 10K low-scoring sentences were chosen at random. The original 20K sentences were run through a gene/protein name tagger, and the results were modified manually to reflect a wide definition of gene/protein names subject to a specificity constraint, a rule that required the tagged entities to refer to specific entities. Each sentence in GENETAG was annotated with acceptable alternatives to the gene/protein names it contained, allowing for partial matching with semantic constraints. Semantic constraints are rules requiring the tagged entity to contain its true meaning in the sentence context. Application of these constraints results in a more meaningful measure of the performance of an NER system than unrestricted partial matching. CONCLUSION: The annotation of GENETAG required intricate manual judgments by annotators which hindered tagging consistency. The data were pre-segmented into words, to provide indices supporting comparison of system responses to the "gold standard". However, character-based indices would have been more robust than word-based indices. GENETAG Train, Test and Round1 data and ancillary programs are freely available at ftp://ftp.ncbi.nlm.nih.gov/pub/tanabe/GENETAG.tar.gz. A newer version of GENETAG-05, will be released later this year.

Animals↗

PubMed2005

A strategy for assigning new concepts in the MEDLINE database.

The MeSH indexing done in MEDLINE is engineered by humans. Humans define the MeSH concepts and human indexers assign MeSH terms to MEDLINE records. Methods have been designed in an attempt to assign MeSH terms to MEDLINE documents automatically with some success. Methods have also been designed to locate useful phrases as potential concepts for indexing. However, little work has been done on the problem of how one might automatically index with the concepts represented by such phrases. Here we examine this issue and present a method for such indexing.

Abstracting and Indexing↗

PubMed2004

GeneWays: a system for extracting, analyzing, visualizing, and integrating molecular pathway data.

The immense growth in the volume of research literature and experimental data in the field of molecular biology calls for efficient automatic methods to capture and store information. In recent years, several groups have worked on specific problems in this area, such as automated selection of articles pertinent to molecular biology, or automated extraction of information using natural-language processing, information visualization, and generation of specialized knowledge bases for molecular biology. GeneWays is an integrated system that combines several such subtasks. It analyzes interactions between molecular substances, drawing on multiple sources of information to infer a consensus view of molecular networks. GeneWays is designed as an open platform, allowing researchers to query, review, and critique stored information.

Artificial Intelligence↗

PubMed2004

Generation of a large gene/protein lexicon by morphological pattern analysis.

The identification of gene/protein names in natural language text is an important problem in named entity recognition. In previous work we have processed MEDLINE documents to obtain a collection of over two million names of which we estimate that perhaps two thirds are valid gene/protein names. Our problem has been how to purify this set to obtain a high quality subset of gene/protein names. Here we describe an approach which is based on the generation of certain classes of names that are characterized by common morphological features. Within each class inductive logic programming (ILP) is applied to learn the characteristics of those names that are gene/protein names. The criteria learned in this manner are then applied to our large set of names. We generated 193 classes of names and ILP led to criteria defining a select subset of 1,240,462 names. A simple false positive filter was applied to remove 8% of this set leaving 1,145,913 names. Examination of a random sample from this gene/protein name lexicon suggests it is composed of 82% (+/-3%) complete and accurate gene/protein names, 12% names related to genes/proteins (too generic, a valid name plus additional text, part of a valid name, etc.), and 6% names unrelated to genes/proteins. The lexicon is freely available at ftp.ncbi.nlm.nih.gov/pub/tanabe/Gene.Lexicon.

Algorithms↗

PubMed2003

The dimensions of indexing.

Indexing of documents is an important strategy intended to make the literature more readily available to the user. Here we describe several dimensions of indexing that are important if indexing is to be optimal. These dimensions are coverage, predictability, and transparency. MeSH terms and text words are compared in MEDLINE in regard to these dimensions. Part of our analysis consists in applying AdaBoost with decisions trees as the weak learners to estimate how reliably index terms are being assigned and how complex the criteria are by which they are being assigned. Our conclusions are that MeSH terms are more predictable and more transparent than text words.

Abstracting and Indexing↗

PubMed2002

Tagging gene and protein names in biomedical text.

MOTIVATION: The MEDLINE database of biomedical abstracts contains scientific knowledge about thousands of interacting genes and proteins. Automated text processing can aid in the comprehension and synthesis of this valuable information. The fundamental task of identifying gene and protein names is a necessary first step towards making full use of the information encoded in biomedical text. This remains a challenging task due to the irregularities and ambiguities in gene and protein nomenclature. We propose to approach the detection of gene and protein names in scientific abstracts as part-of-speech tagging, the most basic form of linguistic corpus annotation. RESULTS: We present a method for tagging gene and protein names in biomedical text using a combination of statistical and knowledge-based strategies. This method incorporates automatically generated rules from a transformation-based part-of-speech tagger, and manually generated rules from morphological clues, low frequency trigrams, indicator terms, suffixes and part-of-speech information. Results of an experiment on a test corpus of 56K MEDLINE documents demonstrate that our method to extract gene and protein names can be applied to large sets of MEDLINE abstracts, without the need for special conditions or human experts to predetermine relevant subsets. AVAILABILITY: The programs are available on request from the authors.

Abbreviations as Topic↗

PubMed2002

A thematic analysis of the AIDS literature.

Faced with the need for human comprehension of any large collection of objects, a time honored approach has been to cluster the objects into groups of closely related objects. Individual groups are then summarized in some convenient manner to provide a more manageable view of the data. Such methods have been applied to document collections with mixed results. If a hard clustering of the data into mutually exclusive clusters is performed then documents are frequently forced into one cluster when they may contain important information that would also appropriately make them candidates for other clusters. If a soft clustering is used there still remains the problem of how to provide a useful summary of the data in a cluster. Here we introduce a new algorithm to produce a soft clustering of document collections that is based on the concept of a theme. A theme is conceptually a subject area that is discussed by multiple documents in the database. A theme has two potential representations that may be viewed as dual to each other. First it is represented by the set of documents that discuss the subject or theme and second it is also represented by the set of key terms that are typically used to discuss the theme. Our algorithm is an EM algorithm in which the term representation and the document representation are explicit components and each is used to refine the other in an alternating fashion. Upon convergence the term representation provides a natural summary of the document representation (the cluster). We describe how to optimize the themes produced by this process and give the results of applying the method to a database of over fifty thousand PubMed documents dealing with the subject of AIDS. How themes may improve access to a document collection is also discussed.

Acquired Immunodeficiency Syndrome↗

PubMed2002

DNA splice site detection: a comparison of specific and general methods.

In an era when whole organism genomes are being routinely sequenced, the problem of gene finding has become a key issue on the road to understanding. For eukaryotic organisms a large part of locating the genes is accomplished by predicting the likely location of splice sites on a DNA strand. This problem of splice site location has been ap- proached using a number of machine learning or statistical methods tailored more or less specifically to the nature of the problem. Recently large margin classifiers and boosting methods have been found to give improvements over more traditional methods in a number of areas. Here we compare large margin classifiers (SVM and CMLS) and boosted decision trees with the three most common models used for splice site detection (WMM, WAM, and MDT). We find that the newer methods compare favorably in all cases and can yield significant improvement in some cases.

Algorithms↗

PubMed2002

Automatic extraction of gene and protein synonyms from MEDLINE and journal articles.

Genes and proteins are often associated with multiple names, and more names are added as new functional or structural information is discovered. Because authors often alternate between these synonyms, information retrieval and extraction benefits from identifying these synonymous names. We have developed a method to extract automatically synonymous gene and protein names from MEDLINE and journal articles. We first identified patterns authors use to list synonymous gene and protein names. We developed SGPE (for synonym extraction of gene and protein names), a software program that recognizes the patterns and extracts from MEDLINE abstracts and full-text journal articles candidate synonymous terms. SGPE then applies a sequence of filters that automatically screen out those terms that are not gene and protein names. We evaluated our method to have an overall precision of 71% on both MEDLINE and journal articles, and 90% precision on the more suitable full-text articles alone

Electronic Data Processing↗

PubMedDate not supplied

Automatically identifying gene/protein terms in MEDLINE abstracts.

MOTIVATION: Natural language processing (NLP) techniques are used to extract information automatically from computer-readable literature. In biology, the identification of terms corresponding to biological substances (e.g., genes and proteins) is a necessary step that precedes the application of other NLP systems that extract biological information (e.g., protein-protein interactions, gene regulation events, and biochemical pathways). We have developed GPmarkup (for "gene/protein-full name mark up"), a software system that automatically identifies gene/protein terms (i.e., symbols or full names) in MEDLINE abstracts. As a part of marking up process, we also generated automatically a knowledge source of paired gene/protein symbols and full names (e.g., LARD for lymphocyte associated receptor of death) from MEDLINE. We found that many of the pairs in our knowledge source do not appear in the current GenBank database. Therefore our methods may also be used for automatic lexicon generation. RESULTS: GPmarkup has 73% recall and 93% precision in identifying and marking up gene/protein terms in MEDLINE abstracts. AVAILABILITY: A random sample of gene/protein symbols and full names and a sample set of marked up abstracts can be viewed at http://www.cpmc.columbia.edu/homepages/yuh9001/GPmarkup/. Contact. hy52@columbia.edu. Voice: 212-939-7028; fax: 212-666-0140.

Abstracting and Indexing↗