Knowledge extraction from biological data

Our current capacity to generate high volumes of biological data through the use of experimental technologies such as genomics, transcriptomics, proteomics or metabolomics (to name just a few) is changing the way in which research is performed across all the biosciences. However, the effectiveness of these technologies is constrained by the capabilities of the analysis methods applied to this data. While great progress has been done throughout the years in this area, most work in the analysis of biological data tends to focus almost exclusively on the core data mining tasks and far less in knowledge discovery aspects. This issue constrains the process of discovering of new biological knowledge that, after all, is the ultimate goal of data-driven biology. For the last decade we have developed methodologies for the analysis of biological data where knowledge discovery is central. The biological problems we tackled challenged our analysis methods in ways that drove us to create superior mining and knowledge discovery algorithms and methodologies. The aim of the talk is to show how bioinformatics has been a driver for innovation in data mining and knowledge discovery, and present our vision for the next generation of knowledge intensive biological data mining methods.