Fazel Famili : Data integration and validation in life sciences
Jeudi 6 septembre 2012 LIRMM
Fazel Famili, Ottawa, Ontario, Canada
Title: Data integration and validation in life sciences
Abstract: Development of novel post-genomics data analysis methods are attributed to the completion of human genome and many other species. The real challenge has been to develop computational methods to first integrate these abundant and heterogeneous omics data for investigation, and to ultimately validate discoveries from various paradigms of biological studies (e.g. genomics and proteomics). To understand the real value of this huge amount of data from each paradigm, several attempts have been made among which are: (i) direct integration of numeric/symbolic data, (ii) tandem analysis of multiple data sets and (iii) integration of discovered knowledge for the purpose of validation and post-processing of patterns and biological signatures. While the first seems to be ideal, there are many challenges that are not trivial, among which are level of data granularity and the problem of one-to-one mapping of data objects. The third paradigm, which involves some manual work, seems to be more effective in generating some novel results including stronger confidence on target discoveries and unraveling global regulatory mechanisms. In this talk we briefly describe the later paradigm and provide details of two case studies. In the first case, markers identified from one genomics data set are validated using proteomics data from the same source. In the second case, patterns discovered in human proteomics data are validated using public gene expression data in order to identify the most promising drug targets. In both cases interesting patterns are revealed from the integrated analysis of biological data.
Bio: Dr. Fazel Famili is a Group Leader, Principal Research Scientist and a leading data mining expert working at the Institute for Information Technology (IIT) of the National Research Council of Canada, where he has been working for the past 27 years. Prior to joining NRC, he worked in industry for 3 years. Dr. Famili has been actively involved in the field of Artificial Intelligence, Data Mining and Bioinformatics and successful application of these technologies. He has a strong data mining and bioinformatics team within IIT that is currently engaged in unique research and development in data mining for Life Sciences. His research has been on data mining, machine learning and bioinformatics and their applications to real world problems in various data rich environments, such as semiconductor manufacturing, aerospace and life sciences. Dr. Famili has edited two books, has published over 50 articles in the area of data mining and AI and has a US data mining patent. He has organized several workshops and has been involved in a number of data mining and AI conferences (e.g. ECAI, ECML/PKDD, AAAI, ICML, and KDD) and has extensive collaboration with a number of Institutes in Canada, Europe, Far East and South America. He is Founding Editor-in-Chief of the Intelligent Data Analysis (a bi-monthly refereed scientific journal established in 1996 which is published by the IOS Press in the Netherlands http://www.iospress.nl/journal/intelligent-data- analysis/), and an adjunct professor at the School of Electrical Engineering and Computer Science, at the University of Ottawa.
Last update on 19/06/2013