Tutorials: Tutorial I: HPC and big data analytics in biomedical informatics

With the increasing number and sophistication of biomedical instruments and data generation devices, there is a tremendous need for advanced data analytics tools to extract useful knowledge out of the collected data. This includes advanced sequencing technologies responsible for the generation of massive amounts of genomics and bioinformatics data as well as wearable devices and Internet-of-Things (IoT) systems responsible for collecting different types of health and mobility data. The currently available data is not only enormous in size but it also exhibits all the features of big data systems with a high degree of variability, velocity and veracity. Such big data systems in the biomedical domain represent great challenges as well as unlimited opportunities to advance biomedical research. Developing innovative data analysis and mining techniques along with clever parallel computational methods to implement them will be critical in efficiently meeting those challenges and take full advantage of the potential opportunities. In particular, the use of graph modeling and network analysis as the backbone of the big data analytics algorithms promises to play an important role in developing data-driven decision support systems and advancing biomedical research. This tutorial focuses on the development of innovative big data analytics tools using graph modeling and network analysis along with how to effectively utilize HPC and private/public clouds in implementing the proposed tools. Case studies illustrating how the tools were used to analyze data associated with infectious diseases that led to new biological discoveries will be presented. The tutorial will also address how mobility data collected from wearable devices can be used to assess health levels and predict potential health hazards. We hope that the proposed methodology and presented results will represent an empirical evidence that the integration of biomedical informatics, HPC and graph-based big data analytics will provide researchers an invaluable tool in advancing the next generation of biomedical research.