Tools
Wednesday, 25th May, 2011, 11:00
An integrative approach for epigenetic data analysis
"High-Throughput sequencing (HTS) has revolutionized the study of gene regulation and expression.
However, there is a strong need for methods that facilitate the integration of multiple datasets to build predictive models. We present a computational framework to analyze and integrate epigenetic data, which allows the development of predictive models of gene regulation. Within this framework, we provide tools to carry out analysis of HTS data from DNA-protein binding, RNA-protein binding and RNA expression assays.
In particular, we have developed a method that can effectively characterize significant changes in epigenetic patterns genome-wide, including promoters, enhancers and genic regions. Furthermore, we provide a tool for building predictive models based on Machine Learning (ML) from multiple datasets. Using the published datasets, we show that our ML methodology allows us to predict the expression change from chromatin properties with 95% accuracy. Additionally, our tools allow the integration of a variety of input datasets and the application of many different ML methods. We finally discuss how this computational framework can be applied to the study of the epigenetic changes in cancer."
Speaker: Eduardo Eyras - Computational Genomics, UPF
Room 473.10 PRBB
