Upcoming Events

Hamid Bolouri   Health Sciences K-069, 2005-09-28 13:30:00 - 2005-09-28 14:30:00

Hamid Bolouri works at the Institue of Systems Biology. I don't know exactly what his talk will cover, but here is a blurb from his web page: In the Bolouri lab, the following disciplines are represented: software engineering systems analysis pattern recognition and classification dynamical systems theory metabolic control analysis The Bolouri lab specializes in the design and application of pattern recognition and adaptive systems and exploits this expertise to: reverse-engineer the computational principles underlying cellular processes develop tools and techniques for modeling and analysis of experimental data at three levels: individual genes network modules whole networks A current focus of the Bolouri lab is to extend and apply the methodologies and software tools it originally developed for reverse engineering sea urchin genetic networks to the large volumes of data for other organisms available in-house at the ISB. The underlying principle is to analyze data from a wide variety of experimental sources looking for consensus among the multiple pieces of evidence. This is an iterative cycle that spirals up toward a clearer understanding of the biological system as a whole. In order to merge observations from different types of experiments automatically, the lab is using the Systems Biology Markup Language, the Systems Biology Workbench, and several software tools developed at the ISB. The aim is to develop a toolkit for Modeling and Analysis of Genetic Regulatory Networks (MAGNET) with standardized component interfaces that facilitate communication between packages, comparison of analysis results, and collective constraint satisfaction.

BioC2005

 

Combining Multiple Diagnostic Tests with NonParametric Transformation Models for Classifying Censored Event Times   FHCRC Arnold Building M1-A307, 2005-09-28 12:00:00 - 2005-09-28 13:00:00

Recent advancement in technology promises to yield a multitude of tests for disease diagnosis and prognosis. When there are multiple sources of information available, it is often of interest to construct a composite score that can provide better classification accuracy than an individual test. In this paper we consider robust procedures for optimally combining tests when test results are measured prior to disease onset and the disease status evolves over time. The most commonly used approach for combining tests to detect subsequent disease status is to fit a proportional hazards model (Cox, 1972) and use the estimated risk score. However, simulation studies suggested that such a risk score may have poor accuracy when the proportional hazards assumption fails. We propose the use of a nonparametric transformation model (Han, 1987) as a working model to derive an optimal composite score with theoretical justification. We demonstrate that the proposed score is the optimal score when the model holds and is optimal ``on average'' among linear scores even if the model fails. Time-dependent sensitivity, specificity and ROC functions are used to quantify the accuracy of the resulting composite score. We provide consistent and asymptotically Gaussian estimators of these accuracy measures. A simple model-free resampling procedure is proposed to obtain all consistent variance estimators. We illustrate the new proposals with simulation studies and an analysis of a breast cancer gene-expression dataset.

Progress in High-Resolution Modeling of Protein Structure and Interactions   Main Campus: HSB K-069, 2005-10-19 13:30:00 - 2005-10-19 13:30:00

Our research is focused on the prediction and design of protein structures, protein folding mechanisms, and protein protein interactions. Our approach is to use experiments to understand the fundamental principles underlying these problems, to develop simple computational models based on these insights, and to test the models through structure prediction and design. A particularly exciting recent success with this approach was the development of the ROSETTA method for ab initio protein structure prediction, which produced de novo structure predictions of unprecedented accuracy in the recent CASP4 international blind test of protein structure prediction methods. We are currently working to appply these methods to the interpretation of genome sequence information.