Neil F. Abernethy

RiboWeb: an ontology-based system for collaborative molecular biology

RiboWeb is an online data resource for the ribosome, a vital cellular apparatus. It contains a large knowledge base of relevant published data and computational modules that can process this data to test hypotheses about the ribosomes structure.

Comparison of Electronic Data Capture (EDC) with the Standard Data Capture Method for Clinical Trial Data

Background Traditionally, clinical research studies rely on collecting data with case report forms, which are subsequently entered into a database to create electronic records. Although well established, this method is time-consuming and error-prone. This study compares four electronic data capture (EDC) methods with the conventional approach with respect to duration of data capture and accuracy. It was performed in a West African setting, where clinical trials involve data collection from urban, rural and often remote locations. Methodology/Principal Findings Three types of commonly available EDC tools were assessed in face-to-face interviews; netbook, PDA, and tablet PC. EDC performance during telephone interviews via mobile phone was evaluated as a fourth method. The Graeco Latin square study design allowed comparison of all four methods to standard paper-based recording followed by data double entry while controlling simultaneously for possible confounding factors such as interview order, interviewer and interviewee. Over a study period of three weeks the error rates decreased considerably for all EDC methods. In the last week of the study the data accuracy for the netbook (5.1%, CI95%: 3.5–7.2%) and the tablet PC (5.2%, CI95%: 3.7–7.4%) was not significantly different from the accuracy of the conventional paper-based method (3.6%, CI95%: 2.2–5.5%), but error rates for the PDA (7.9%, CI95%: 6.0–10.5%) and telephone (6.3%, CI95% 4.6–8.6%) remained significantly higher. While EDC-interviews take slightly longer, data become readily available after download, making EDC more time effective. Free text and date fields were associated with higher error rates than numerical, single select and skip fields. Conclusions EDC solutions have the potential to produce similar data accuracy compared to paper-based methods. Given the considerable reduction in the time from data collection to database lock, EDC holds the promise to reduce research-associated costs. However, the successful implementation of EDC requires adjustment of work processes and reallocation of resources.

A tuberculosis outbreak in a private-home family child care center in San Francisco, 2002 to 2004.

BACKGROUND. Child care facilities are well known as sites of infectious disease transmission, and California child care facility licensure requirements include annual tuberculosis (TB) screening for on-site adults. In April 2004, we detected an adult with TB living in a private-home family child care center (child care center A). METHODS. We reviewed patient medical records and conducted a contact investigation. The investigation included all persons at the child care center, the workplace and leisure contacts of the adult patient with TB, and the household contacts of secondary case patients. Contact names were obtained through patient interviews. A positive tuberculin skin test result was defined as induration of ≥5 mm. DNA fingerprints of Mycobacterium tuberculosis isolates were analyzed. Outbreak cases were those that had matching DNA fingerprint patterns or were linked epidemiologically, if DNA fingerprint results were not available. RESULTS. Between August 2002 and July 2004, we detected 11 outbreak cases, including 9 (82%) among children (<18 years of age). All 11 outbreak patients lived or were cared for at child care center A. The 9 pediatric TB patients were young (<7 years of age), United States-born children of foreign-born parents, and 4 (44%) had positive cultures for M tuberculosis. Including isolates recovered from the 2 adult patients, all 6 M tuberculosis isolates shared identical, 7-band, DNA fingerprint patterns. The contact investigation identified 3 (33%) of the 9 pediatric cases; 2 (22%) presented with illness and 4 (44%) were detected by primary care providers during routine TB screening. Excluding case subjects, 36 (54%) of 67 named contacts had latent TB infection. CONCLUSIONS. Provider adherence to locally adapted pediatric TB screening recommendations proved critical to outbreak control. TB screening compliance by the child care center and more aggressive source-case investigation by the TB program might have prevented or abated this large pediatric TB outbreak.

Sophia: a flexible, Web-based knowledge server

Sophia is a frame based knowledge server built on a commercial relational database system. The system is thus simple and inexpensive, while also providing some advanced functionality. Sophia is accessible to users through the Web or to client applications through an API.

Microevolution of the Direct Repeat Locus of Mycobacterium tuberculosis in a Strain Prevalent in San Francisco

ABSTRACT We describe a microevolutionary event of a prevalent strain of Mycobacterium tuberculosis that caused two outbreaks in San Francisco. During the second outbreak, a direct variable repeat was lost. We discuss the mechanisms of this change and the implications of analyzing multiple genetic loci in this context.

Supervised learning for the automated transcription of spacer classification from spoligotype films

BackgroundMolecular genotyping of bacteria has revolutionized the study of tuberculosis epidemiology, yet these established laboratory techniques typically require subjective and laborious interpretation by trained professionals. In the context of a Tuberculosis Case Contact study in The Gambia we used a reverse hybridization laboratory assay called spoligotype analysis. To facilitate processing of spoligotype images we have developed tools and algorithms to automate the classification and transcription of these data directly to a database while allowing for manual editing.ResultsFeatures extracted from each of the 1849 spots on a spoligo film were classified using two supervised learning algorithms. A graphical user interface allows manual editing of the classification, before export to a database. The application was tested on ten films of differing quality and the results of the best classifier were compared to expert manual classification, giving a median correct classification rate of 98.1% (inter quartile range: 97.1% to 99.2%), with an automated processing time of less than 1 minute per film.ConclusionThe software implementation offers considerable time savings over manual processing whilst allowing expert editing of the automated classification. The automatic upload of the classification to a database reduces the chances of transcription errors.

Assessing network characteristics of cancer associated genes in metabolic and signaling networks

In the post-genome era, high-throughput experimental methods have elucidated many of the complex interactions in metabolic, regulatory, and signal transduction pathways. Graph theoretic methods have been broadly applied to study properties of these interactions. Here we explore the relationship between network properties of genes and their implication in cancer etiology. We extract pathway interactions from the Kyoto Encyclopedia of Genes and Genomes (KEGG) to create global signaling and metabolic networks. Using a generalized linear model, we evaluate the predictive power of centrality measures and clustering coefficient. We then apply a random-walk algorithm to discover communities enriched with cancer-associated genes. Our findings show cancer genes in metabolic and signaling networks exhibit significant topological differences considering degree, clustering coefficient, and community cohesiveness; and these features demonstrate greater predictive power in signaling networks. These results support an empirical basis for algorithms using similar network-based measures to prioritize disease genes or predict disease states.

Linking information systems for HIV care and research in Kenya

The provision of HIV care in developing countries may involve complex and overlapping resources; including government-run facilities non-governmental organization (NGO) or international non-governmental organization (INGO) supported services and research affiliated clinics. These resources are often motivated and funded by distinct health priorities and as a result, standards for clinical data representation and exchange are rare and data management systems are often redundant. Open-source systems such as OpenMRS and OpenClinica provide an opportunity to leverage available systems to improve standards and increase interoperability. Nevertheless, continuity of care and data sharing between these systems remains a challenge, particularly in populations with changing health needs, high mobility, and inconsistent access to health resources. As a prerequisite to improving interoperability between systems, use cases for clinical information exchange are first identified. We then characterize data models from nine clinical information systems, standards, and ontologies pertinent to HIV clinical care and research in Kenya. The data fields commonly used as patient identifiers are summarized, including name, date of birth, family relations and location. Finally, we present a prototype ontology to describe data standards and to enable mapping between data elements in diverse information systems.