Yardena Peres

Selecting anti-HIV therapies based on a variety of genomic and clinical factors

Motivation: Optimizing HIV therapies is crucial since the virus rapidly develops mutations to evade drug pressure. Recent studies have shown that genotypic information might not be sufficient for the design of therapies and that other clinical and demographical factors may play a role in therapy failure. This study is designed to assess the improvement in prediction achieved when such information is taken into account. We use these factors to generate a prediction engine using a variety of machine learning methods and to determine which clinical conditions are most misleading in terms of predicting the outcome of a therapy. Results: Three different machine learning techniques were used: generative–discriminative method, regression with derived evolutionary features, and regression with a mixture of effects. All three methods had similar performances with an area under the receiver operating characteristic curve (AUC) of 0.77. A set of three similar engines limited to genotypic information only achieved an AUC of 0.75. A straightforward combination of the three engines consistently improves the prediction, with significantly better prediction when the full set of features is employed. The combined engine improves on predictions obtained from an online state-of-the-art resistance interpretation system. Moreover, engines tend to disagree more on the outcome of failure therapies than regarding successful ones. Careful analysis of the differences between the engines revealed those mutations and drugs most closely associated with uncertainty of the therapy outcome. Availability: The combined prediction engine will be available from July 2008, see http://engine.euresist.org Contact: rosen@il.ibm.com

Predicting the Response to Combination Antiretroviral Therapy: Retrospective Validation of geno2pheno-THEO on a Large Clinical Database

BACKGROUND Expert-based genotypic interpretation systems are standard methods for guiding treatment selection for patients infected with human immunodeficiency virus type 1. We previously introduced the software pipeline geno2pheno-THEO (g2p-THEO), which on the basis of viral sequence predicts the response to treatment with a combination of antiretroviral compounds by applying methods from statistical learning and the estimated potential of the virus to escape from drug pressure. METHODS We retrospectively validated the statistical model used by g2p-THEO in approximately 7600 independent treatment-sequence pairs extracted from the EuResist integrated database, ranging from 1990 to 2007. Results were compared with the 3 most widely used expert-based interpretation systems: Stanford HIVdb, ANRS, and Rega. RESULTS The difference in receiver operating characteristic curves between g2p-THEO and expert-based approaches was significant (P < .001; paired Wilcoxon test). Indeed, at 80% specificity, g2p-THEO found 16.2%-19.8% more successful regimens than did the expert-based approaches. The increased performance of g2p-THEO was confirmed in a 2001-2007 data set from which most obsolete therapies had been removed. CONCLUSION Finding drug combinations that increase the chances of therapeutic success is the main reason for using decision support systems. The present analysis of a large data set derived from clinical practice demonstrates that g2p-THEO solves this task significantly better than state-of-the-art expert-based systems. The tool is available at http://www.geno2pheno.org.

Comparison of classifier fusion methods for predicting response to anti HIV-1 therapy.

Background Analysis of the viral genome for drug resistance mutations is state-of-the-art for guiding treatment selection for human immunodeficiency virus type 1 (HIV-1)-infected patients. These mutations alter the structure of viral target proteins and reduce or in the worst case completely inhibit the effect of antiretroviral compounds while maintaining the ability for effective replication. Modern anti-HIV-1 regimens comprise multiple drugs in order to prevent or at least delay the development of resistance mutations. However, commonly used HIV-1 genotype interpretation systems provide only classifications for single drugs. The EuResist initiative has collected data from about 18,500 patients to train three classifiers for predicting response to combination antiretroviral therapy, given the viral genotype and further information. In this work we compare different classifier fusion methods for combining the individual classifiers. Principal Findings The individual classifiers yielded similar performance, and all the combination approaches considered performed equally well. The gain in performance due to combining methods did not reach statistical significance compared to the single best individual classifier on the complete training set. However, on smaller training set sizes (200 to 1,600 instances compared to 2,700) the combination significantly outperformed the individual classifiers (p<0.01; paired one-sided Wilcoxon test). Together with a consistent reduction of the standard deviation compared to the individual prediction engines this shows a more robust behavior of the combined system. Moreover, using the combined system we were able to identify a class of therapy courses that led to a consistent underestimation (about 0.05 AUC) of the system performance. Discovery of these therapy courses is a further hint for the robustness of the combined system. Conclusion The combined EuResist prediction engine is freely available at http://engine.euresist.org.

Prediction of Response to Antiretroviral Therapy by Human Experts and by the EuResist Data-Driven Expert System (the EVE study)

The EuResist expert system is a novel data‐driven online system for computing the probability of 8‐week success for any given pair of HIV‐1 genotype and combination antiretroviral therapy regimen plus optional patient information. The objective of this study was to compare the EuResist system vs. human experts (EVE) for the ability to predict response to treatment.

Information technology for healthcare transformation

Rising costs, decreasing quality of care, diminishing productivity, and increasing complexity have all contributed to the present state of the healthcare industry. The interactions between payers (e.g., insurance companies and health plans) and providers (e.g., hospitals and laboratories) are growing and are becoming more complicated. The constant upsurge in and enhanced complexity of diagnostic and treatment information has made the clinical decision-making process more difficult. Medical transaction charges are greater than ever. Population-specific financial requirements are increasing the economic burden on the entire system. Medical insurance and identity theft frauds are on the rise. The current lack of comparative cost analytics hampers systematic efficiency. Redundant and unnecessary interventions add to medical expenditures that add no value. Contemporary payment models are antithetic to outcome-driven medicine. The rate of medical errors and mistakes is high. Slow inefficient processes and the lack of best practice support for care delivery do not create productive settings. Information technology has an important role to play in approaching these problems. This paper describes IBM Researchs approach to helping address these issues, i.e., the evidence-based healthcare platform.

Browsing and Editing XML Schema Documents with an Interactive Editor

With the advent of the web there has been a great demand for data interchange between existing applications using internet infrastructure and also between newer web services applications. The W3C XML standard is becoming the internet data interchange format, even though the initial XML standard was not well suited to this. The XML Schema recommendation which added more rigorous data structuring and data typing provides much better support for defining such data centric XML documents. While data centric XML documents are typically produced by applications there are still a wide range of situations where manual document creation is required. This paper presents an XML editor design which supports the creation of XML Schema based documents. We show how an interface that uses a tight coupling between grammar and content views facilitates the rapid creation of data centric documents. We show how our interface supports browsing and editing of documents where XML schema specific extensions such as subtyping are present. Our design is realised in the Xeena for Schema tool which we demonstrate.

Xeena for Schema: Creating XML Data with an Interactive Editor

With the advent of the web there has been a great demand for data interchange between existing applications using internet infrastructure and also between newer web services applications. The W3C XML standard is becoming the internet data interchange format. Such XML data is typically produced by applications. However during application development and maintenance there remains a significant need for manual creation, editing and browsing of XML data by application and system developers. XML editors can fill this need. This paper presents an interactive XML editor design. We show how an interface that uses a tight coupling between grammar and content views facilitates the rapid creation of data centric documents. Our design is realised in the Xeena for Schema tool which we demonstrate. Xeena for Schema supports the latest version of XML, XML Schema, which offers better support for data oriented applications.

Efficacy of Antiretroviral Therapy Switch in HIV-Infected Patients: A 10- year Analysis of the EuResist Cohort

Introduction: Highly active antiretroviral therapy (HAART) has been shown to be effective in many recent trials. However, there is limited data on time trends of HAART efficacy after treatment change. Methods: Data from different European cohorts were compiled within the EuResist Project. The efficacy of HAART defined by suppression of viral replication at 24 weeks after therapy switch was analyzed considering previous treatment modifications from 1999 to 2008. Results: Altogether, 12,323 treatment change episodes in 7,342 patients were included in the analysis. In 1999, HAART after treatment switch was effective in 38.0% of the patients who had previously undergone 1–5 therapies. This figure rose to 85.0% in 2008. In patients with more than 5 previous therapies, efficacy rose from 23.9 to 76.2% in the same time period. In patients with detectable viral load at therapy switch, the efficacy rose from 23.3 to 66.7% with 1–5 previous treatments and from 14.4 to 55.6% with more than 5 previous treatments. Conclusion: The results of this large cohort show that the outcome of HAART switch has improved considerably over the last years. This result was particularly observed in the context after viral rebound. Thus, changing HAART is no longer associated with a high risk of treatment failure.

BioMIMS - SOA Platform for Research of Rare Hereditary Diseases

Introduction: BioMIMS is an award-winning* platform that realizes our vision for how information technologies can support the research of rare hereditary diseases. A disease is considered rare when it affects only a small percentage of the population, most rare diseases are genetic. Researching rare hereditary diseases imposes several significant challenges for dedicated informatics tools. Collaboration is a critical element in the research process, and partnerships among research centers are essential to ensure advances in understanding and treating these diseases. Because a single research center usually lacks sufficient data for conducting meaningful research, data must be shared among partners. Consequently, the first challenge of the underlying platform is to merge information gathered from dispersed hospitals, research centers, clinics, labs and other facilities. Additionally, various data types such as clinical, genomic, imaging data, and pedigrees, must be combined to create a comprehensive disease view. Obtaining and visualizing all the available data is the first step toward gaining insights through analytics. Research questions cannot always be known in advance and tend to change over time; therefore, an effective platform should be able to integrate various data mining and analytic algorithms. Platform Architecture: We designed BioMIMS based on Service Oriented Architecture (SOA) principles. The BioMIMS architecture is composed of a rich set of fine-grained services decoupled by a central bus. The bus orchestrates the existing services according to predefined workflows, following the Message Oriented Middleware (MOM) concept. Message queues provide temporary storage when the destination service is busy or not connected. All the messages running through the bus and handled by the different services are based on approved industry standards. Medical images are uploaded to and retrieved from the appropriate system service in DICOM v3.0 format, while clinical and family history data are uploaded and retrieved according to HL7 v2.x and HL7 v3 Family History standards. Patients may have different patient identifiers in different systems. To ensure the correct identification of patients and their data in a standard manner, BioMIMS supports IHE Patient Identifier Cross-Reference (PIX) and Patient Demographic Query (PDQ) transactions. To enable interoperability at the cross-hospital and regional levels, metadata are extracted from the stored information according to IHE Cross-enterprise Document Sharing (XDS/XDS-I) Profiles. A SOA approach, based on standard interfaces has numerous benefits, namely: a) a flexible architecture that allows easy integration of new services, creation of new workflows by reusing existing ones, and easy integration with existing applications, b) inherent scalability, c) speedy custom application development that reduces total IT costs. Platform Validation: BioMIMS was validated by researchers of the Rizzoli Orthopaedic Institute (IOR). IOR researchers uploaded and investigated data for two different orthopedic diseases -- Multiple Osteochondromas (MO) and Osteogenesis Imperfecta (OI) The researchers confirmed several presumed insights for the available data using BioMIMS. Moreover, a few interesting points arose during the validation process, helping to determine goals for future studies. * BioMIMS is a 2010 Computerworld Honors Laureate and an IBM Smarter Planet Reference Account.

A Framework Registration Language

In order to maintain a framework generic, one must provide means that will allow the framework users (application developers) means to extend and adapt it to their specific needs. This implies that some kind of registration mechanism is needed in order to keep the framework neutral of any specific application. The registration mechanism is needed in order to allow the framework to become aware and be able to control the user objects/extensions. To overcome this need we have introduced a framework registration language that allows developers to register and maintain their framework extensions. Moreover the registration language allows the developers to render dialog controls and many UI related behaviors, such as drag/drop and menu items, without any code. As an outcome of these capabilities the language enabled developers to develop fast prototypes based on the registration language only.