Workshop: Solving Big Data Challenges in Modern Science through Statistical Modelling

Organisers: Mayetri Gupta (University of Glasgow), Surajit Ray (University of Glasgow), Indranil Mukhopadhyay (Indian Statistical Institute) Debasis Sengupta (Indian Statistical Institute)

Driven by the statistical challenges of "Big Data" analysis in modern scientific applications, the goal of this workshop is to shape future directions in the area of big data modelling and analysis, consolidating the strengths of statisticians from the UK and India as well as training future researchers. Due to increases in computational power, statistical techniques are now used to fit increasingly complex models to increasingly large datasets arising from a variety of fields, ranging from medical imaging, genomic and genetic high-throughput assays, to remote sensing, power grids, and social networks. In recent years, there has been a huge upsurge in data collection in genomics and medicine- vast databases have come into existence for recording every piece of biological information ever collected about individual locations in the human genome (as well as many other species). At the same time, there have been massive improvements in the scope and accuracy of measuring instruments, that have led to a rich record of climate and environmental exposure data over a wide geographical area. However, for all this information to be of actual clinical use- for example, in risk prediction for cancer or a variety of other diseases and genetic conditions- there need to be the development of statistical models and methods that are appropriate to handle the complexity and diversity of the types of data as well as be computationally feasible to implement in real time.

The new applications pose challenges in computation, estimation, and especially in statistically sound inference techniques for high dimensional data. Statistical techniques developed for analysing big data arising from different application areas are quite diverse, yet the underlying challenges are often very similar; but there are few venues where statisticians working on different topic areas get a chance to exchange their ideas and expertise in sufficient depth. In the workshop, we will focus on bringing together ideas on big data analysis challenges from three major application areas:

• Genetics, Genomics and Biomedical Applications
• Environmental issues
• Energy, infrastructure, and emerging technologies

Building cross-connections through the workshop, through the exchange of statistical and mathematical ideas, techniques and tools, the end-goal would be to consolidate these into a novel synthesis of methodologies leading to new breakthroughs in solving scientific problems from a multi-pronged and multi-disciplinary statistical perspective.

Detailed information on the location, accommodation and travel is available at the official ICMS workshop page.

More news and updates will be posted here shortly