Retrieval with gene queries

Ferret: a sentence-based literature scanning system

BACKGROUND

The rapid pace of bioscience research makes it very challenging to track relevant articles in one's area of interest. MEDLINE, a primary source for biomedical literature, offers access to more than 20 million citations with three-quarters of a million new ones added each year. Thus it is not surprising to see active research in building new document retrieval and sentence retrieval systems. We present Ferret, a prototype retrieval system, designed to retrieve and rank sentences (and their documents) conveying gene-centric relationships of interest to a scientist. The prototype has several features. For example, it is designed to handle gene name ambiguity and perform query expansion. Inputs can be a list of genes with an optional list of keywords. Sentences are retrieved across species but the species discussed in the records are identified. Results are presented in the form of a heat map and sentences corresponding to specific cells of the heat map may be selected for display. Ferret is designed to assist bio scientists at different stages of research from early idea exploration to advanced analysis of results from bench experiments.

RESULTS

Three live case studies in the field of plant biology are presented related to Arabidopsis thaliana. The first is to discover genes that may relate to the phenotype of open immature flower in Arabidopsis. The second case is about finding associations reported between ethylene signaling and a set of 300+ Arabidopsis genes. The third case is on searching for potential gene targets of an Arabidopsis transcription factor hypothesized to be involved in plant stress responses. Ferret was successful in finding valuable information in all three cases. In the first case the bZIP family of genes was identified. In the second case sentences indicating relevant associations were found in other species such as potato and jasmine. In the third sentences led to new research questions about the plant hormone salicylic acid.

CONCLUSIONS

Ferret successfully retrieved relevant gene-centric sentences from PubMed records. The three case studies demonstrate end user satisfaction with the system.

Functionally enigmatic genes: a case study of the brain ignorome

What proportion of genes with intense and selective expression in specific tissues, cells, or systems are still almost completely uncharacterized with respect to biological function? In what ways do these functionally enigmatic genes differ from well-studied genes? To address these two questions, we devised a computational approach that defines so-called ignoromes. As proof of principle, we extracted and analyzed a large subset of genes with intense and selective expression in brain. We find that publications associated with this set are highly skewed--the top 5% of genes absorb 70% of the relevant literature. In contrast, approximately 20% of genes have essentially no neuroscience literature. Analysis of the ignorome over the past decade demonstrates that it is stubbornly persistent, and the rapid expansion of the neuroscience literature has not had the expected effect on numbers of these genes. Surprisingly, ignorome genes do not differ from well-studied genes in terms of connectivity in coexpression networks. Nor do they differ with respect to numbers of orthologs, paralogs, or protein domains. The major distinguishing characteristic between these sets of genes is date of discovery, early discovery being associated with greater research momentum--a genomic bandwagon effect. Finally we ask to what extent massive genomic, imaging, and phenotype data sets can be used to provide high-throughput functional annotation for an entire ignorome. In a majority of cases we have been able to extract and add significant information for these neglected genes. In several cases--ELMOD1, TMEM88B, and DZANK1--we have exploited sequence polymorphisms, large phenome data sets, and reverse genetic methods to evaluate the function of ignorome genes.

MeSH: a window into full text for document summarization

MOTIVATION

Previous research in the biomedical text-mining domain has historically been limited to titles, abstracts and metadata available in MEDLINE records. Recent research initiatives such as TREC Genomics and BioCreAtIvE strongly point to the merits of moving beyond abstracts and into the realm of full texts. Full texts are, however, more expensive to process not only in terms of resources needed but also in terms of accuracy. Since full texts contain embellishments that elaborate, contextualize, contrast, supplement, etc., there is greater risk for false positives. Motivated by this, we explore an approach that offers a compromise between the extremes of abstracts and full texts. Specifically, we create reduced versions of full text documents that contain only important portions. In the long-term, our goal is to explore the use of such summaries for functions such as document retrieval and information extraction. Here, we focus on designing summarization strategies. In particular, we explore the use of MeSH terms, manually assigned to documents by trained annotators, as clues to select important text segments from the full text documents.

RESULTS

Our experiments confirm the ability of our approach to pick the important text portions. Using the ROUGE measures for evaluation, we were able to achieve maximum ROUGE-1, ROUGE-2 and ROUGE-SU4 F-scores of 0.4150, 0.1435 and 0.1782, respectively, for our MeSH term-based method versus the maximum baseline scores of 0.3815, 0.1353 and 0.1428, respectively. Using a MeSH profile-based strategy, we were able to achieve maximum ROUGE F-scores of 0.4320, 0.1497 and 0.1887, respectively. Human evaluation of the baselines and our proposed strategies further corroborates the ability of our method to select important sentences from the full texts.

CONTACT

sanmitra-bhattacharya@uiowa.edu; padmini-srinivasan@uiowa.edu.

Information discovery on electronic health records using authority flow techniques

BACKGROUND

As the use of electronic health records (EHRs) becomes more widespread, so does the need to search and provide effective information discovery within them. Querying by keyword has emerged as one of the most effective paradigms for searching. Most work in this area is based on traditional Information Retrieval (IR) techniques, where each document is compared individually against the query. We compare the effectiveness of two fundamentally different techniques for keyword search of EHRs.

METHODS

We built two ranking systems. The traditional BM25 system exploits the EHRs' content without regard to association among entities within. The Clinical ObjectRank (CO) system exploits the entities' associations in EHRs using an authority-flow algorithm to discover the most relevant entities. BM25 and CO were deployed on an EHR dataset of the cardiovascular division of Miami Children's Hospital. Using sequences of keywords as queries, sensitivity and specificity were measured by two physicians for a set of 11 queries related to congenital cardiac disease.

RESULTS

Our pilot evaluation showed that CO outperforms BM25 in terms of sensitivity (65% vs. 38%) by 71% on average, while maintaining the specificity (64% vs. 61%). The evaluation was done by two physicians.

CONCLUSIONS

Authority-flow techniques can greatly improve the detection of relevant information in EHRs and hence deserve further study.

Building a high-quality sense inventory for improved abbreviation disambiguation

Motivation: The ultimate goal of abbreviation management is to disambiguate every occurrence of an abbreviation into its expanded form (concept or sense). To collect expanded forms for abbreviations, previous studies have recognized abbreviations and their expanded forms in parenthetical expressions of bio-medical texts. However, expanded forms extracted by abbreviation recognition are mixtures of concepts/senses and their term variations. Consequently, a list of expanded forms should be structured into a sense inventory, which provides possible concepts or senses for abbreviation disambiguation.

Results: A sense inventory is a key to robust management of abbreviations. Therefore, we present a supervised approach for clustering expanded forms. The experimental result reports 0.915 F1 score in clustering expanded forms. We then investigate the possibility of conflicts of protein and gene names with abbreviations. Finally, an experiment of abbreviation disambiguation on the sense inventory yielded 0.984 accuracy and 0.986 F1 score using the dataset obtained from MEDLINE abstracts.

Availability: The sense inventory and disambiguator of abbreviations are accessible at http://www.nactem.ac.uk/software/acromine/ and http://www.nactem.ac.uk/software/acromine_disambiguation/

Contact:okazaki@chokkan.org

Overview of BioCreative II gene normalization

Background:

The goal of the gene normalization task is to link genes or gene products mentioned in the literature to biological databases. This is a key step in an accurate search of the biological literature. It is a challenging task, even for the human expert; genes are often described rather than referred to by gene symbol and, confusingly, one gene name may refer to different genes (often from different organisms). For BioCreative II, the task was to list the Entrez Gene identifiers for human genes or gene products mentioned in PubMed/MEDLINE abstracts. We selected abstracts associated with articles previously curated for human genes. We provided 281 expert-annotated abstracts containing 684 gene identifiers for training, and a blind test set of 262 documents containing 785 identifiers, with a gold standard created by expert annotators. Inter-annotator agreement was measured at over 90%.

Results:

Twenty groups submitted one to three runs each, for a total of 54 runs. Three systems achieved F-measures (balanced precision and recall) between 0.80 and 0.81. Combining the system outputs using simple voting schemes and classifiers obtained improved results; the best composite system achieved an F-measure of 0.92 with 10-fold cross-validation. A 'maximum recall' system based on the pooled responses of all participants gave a recall of 0.97 (with precision 0.23), identifying 763 out of 785 identifiers.

Conclusion:

Major advances for the BioCreative II gene normalization task include broader participation (20 versus 8 teams) and a pooled system performance comparable to human experts, at over 90% agreement. These results show promise as tools to link the literature with biological databases.

Anni 2.0: a multipurpose text-mining tool for the life sciences

Anni 2.0 provides an ontology-based interface to MEDLINE.

Anni 2.0 is an online tool () to aid the biomedical researcher with a broad range of information needs. Anni provides an ontology-based interface to MEDLINE and retrieves documents and associations for several classes of biomedical concepts, including genes, drugs and diseases, with established text-mining technology. In this article we illustrate Anni's usability by applying the tool to two use cases: interpretation of a set of differentially expressed genes, and literature-based knowledge discovery.

GO for gene documents

Background

Annotating genes and their products with Gene Ontology codes is an important area of research. One approach is to use the information available about these genes in the biomedical literature. The goal in this paper, based on this approach, is to develop automatic annotation methods that can supplement the expensive manual annotation processes currently in place.

Results

Using a set of Support Vector Machines (SVM) classifiers we were able to achieve Fscores of 0.49, 0.41 and 0.33 for codes of the molecular function, cellular component and biological process GO hierarchies respectively. We find that alternative term weighting strategies are not different from each other in performance and feature selection strategies reduce performance. The best thresholding strategy is one where a single threshold is picked for each hierarchy. Hierarchy level is important especially for molecular function and biological process. The cellular component hierarchy stands apart from the other two in many respects. This may be due to fundamental differences in link semantics. This research shows that it is possible to beneficially exploit the hierarchical structures by defining and testing a relaxed criteria for classification correctness. Finally it is possible to build classifiers for codes with very few associated documents but as expected a huge penalty is paid in performance.

Conclusion

The GO annotation problem is complex. Several key observations have been made as for example about topic drift that may be important to consider in annotation strategies.

Adaptive classifiers, topic drifts and GO annotations

Gene annotations with Gene Ontology codes offer scientists important options in their study of genes and their functions. Automatic GO annotation methods have the potential to supplement the intensive manual annotation processes. Annotation approaches using MEDLINE documents are generally two-phased where the first is to annotate documents with GO codes and the second is to annotate gene products via the documents. In this paper we study document annotation with GO codes using a temporal perspective. Specifically, we build adaptive code-specific classifiers. We also study topic drift i.e., changes in the contextual characteristics of annotations over time. We show that topic drift is significant especially in the biological process GO hierarchy. This at least partially explains the particular challenges faced with codes of this hierarchy.

An application of bioinformatics and text mining to the discovery of novel genes related to bone biology

The treatment and management of complex genetic diseases such as osteoporosis can greatly benefit from the integration of relevant research across many different disciplines. We created a text mining tool that analyzes the PubMed literature database and integrates the available genomic information to provide a detailed mapping of the genes and their interrelationships within a particular network such as osteoporosis. The results obtained from our text mining program show that existing genomic data within the PubMed database can effectively be used to predict potentially novel target genes for osteoporosis research that have not previously been reported in the literature. To filter the most significant findings, we developed a ranking system to rate our predicted novel genes. Some of our predicted genes ranked higher than those currently studied, suggesting that they may be of particular interest from a therapeutic standpoint. A preliminary analysis of the current biomedical literature in our research area using our tool suggests that S100A12, as well as a group of SMAD genes previously unstudied in relation to osteoporosis, may be highly relevant to the mechanism of action of bisphosphonates, that the function of osteocytes may be influenced by a family of important interleukins and interleukin-related molecules, and that the FYN oncogene may play an important role in regulating the apoptosis of bone cells in the context of degenerative bone diseases. An evaluation of our tool's predictive ability with an analysis of PubMed literature published before the year 2000 in the area of osteoporosis research shows that many of its top-rated novel target genes from that analysis were later studied and shown to be relevant to osteoporosis in the period between 2000 and 2006. We believe that our tool will be beneficial to researchers in the field of orthopaedics seeking to identify novel target genes in their research area, and it will allow them to delve deeper into the complex interplay between genes, biological systems and diseases.