Multidisciplinary banner Imperial College's strengths lie in multidisciplinary research and innovation. The Institute is all-encompassing, not only developing foundational research in data science methods, technologies and policy, but also in applying data science in all areas of science, engineering, business and medicine.

Applications

Understanding Biology

With the advances in high-throughput experimental technologies, biologists are starting to grapple with massive data sets. Multi-omics data such as genomics, lipidomics, proteomics, transcriptomics and metabolomics data are growing astronomically. These data create a potential bonanza, as they provide systems-level measurements for virtually all types of cellular components in a model organism, yielding unprecedented views of the cellular inner workings. Meanwhile, these data also raise many new research challenges, not only because of their size, but also their increasing complexity. 

At the DSI, we are working with biologists and quantitative scientists including statisticians, mathematicians, engineers and computer scientists, to enable next generation bioinformatics. To make sense of ‘omics’-scale data and get insights from these data that will advance our understanding of the underlying biological mechanisms, different areas including systems biology, statistics, machine learning and high performance computing, are brought together in the DSI to explore the fundamental questions in biology. Further it is expected to lead to practical applications in medicine, drug discovery, and bioengineering.

Economics, finance and value

Data in financial markets is constantly generated and quickly changing, where traditional statistical models quickly become out-of-date, and increasingly high performance real time machine learning algorithms that can adapt and change based on the new data are giving a new edge to trading. This is an active area of research where the DSI and Imperial College’s quantitative financial researchers are working together to make exciting progress. Beyond using data science for exploring competitive advantages in economies, new models for the economy itself are emerging.

For example, money itself is increasingly exchanged digitally, which brings new research questions such as: How to realise a new commercial structure with digital money? Does digital money adoption make a difference? What are the big data implications of digital money? Is it possible to quantify the benefits to governments, corporations and individuals? What are the factors that affect the outcome of a digital money initiative? Going beyond this, data itself could be considered a currency. At the DSI we are beginning to investigate new models for the digital economy as well as supporting data-driven research in economics and finance.

Check out our case study on:

Data-driven Engineering

Engineering and understanding complex systems has always been data intensive, where the main branches of engineering: chemical engineering, materials and earth science and engineering, civil, electrical and mechanical engineering, all utilise simulation and modelling to make advances in technology. With the complexity of computer-based models ever increasing, the simulations on such models are increasingly data-intensive and require a range of numerical methods.

DSI works closely with our colleagues to use data in a broad range of engineering areas including in multiphase flow research, sensor networks, robotics, high-performance computing hardware design such as in FPGAs, aeronautics, and bioengineering. The DSI aims to support engineering research by assisting in the application of big data across engineering disciplines. 

Beyond data driving progress in engineering, engineering also enables the invention of new hardware devices to harvest data and harness the power of data. Computing devices equipped with sensors are today pervasive, where The Internet of Things is a driving force generating data and in-turn pushing forward the forefront of innovations in big data technology. The DSI stays at the forefront of data-driven engineering research - for example, we are actively engaged in developing pervasive sensing technology for healthcare by combining data streams taken from body sensors (EEG, actimetry, gait), environmental (temperature, light, pollution), and behavioural sensing (emotional state, geo-location, consumer behaviour). 

Check out our case study on:

Natural Environment

Where understanding the human body and biological processes is a complex and data-intensive challenge, consider scaling up our understanding of the complexities of nature to the environment around us, from the climate to our planet. Simulation, modelling and knowledge discovery needs to operate at every biological scale extending up to a global scale that not only includes biological scales (from microscopic upwards), but also behavioural complexities of interacting species, geophysical and meteorological systems, and biodiversity and even the spread of disease. 

At the DSI, we are working with environmental researchers to understand how to deal with their data, apply new statistical techniques, and lower the barriers to harnessing complex data in their research by building new tools and training scientists in data science methods. We are collaborating with world-leading research groups and centres, such as Imperial’s own Space and Atmospheric Physics Group, and the Grantham Institute for Climate Change, to help them harness the power of big data in climate prediction, risk mitigation, and natural disaster planning.

Check out our case studies on:

Personalized Medicine

Personalised medicine is about tailoring medical treatment to the individual characteristics of each patient, by classifying individuals into groups that differ in their susceptibility to a particular disease or their response to a specific treatment.  With the amount of data being mined and analysed, it will be easier to identify genetic correlations, identify patterns in patient and population data, identify patient specific patterns and predict physiological conditions, discovering biomarkers that present signs of normal or abnormal processes and provide better patient self-management for enhanced clinical outcomes. The proliferation of data, generated from high-throughput molecular profiling to physiological sensing, offers great opportunities for personalised medicine by offering more precise diagnoses and more effective treatments, as researchers are able to drill down to see what is happening and create more targeted therapies, specifically at the molecular and tissue levels.

At the DSI we are working with Imperial College’s biomedical researchers in the medical school and the worldwide medical research community to build big data technologies to enable personal medicine.  This research includes building the European Translational Knowledge Management and Service (eTRIKS) platform as the gold standard personalised medicine big data research platform for the European medical research community.

Check out our case study on:

Understanding Nature

In recent years, technological advances have dramatically increased the quality and quantity of data available to astronomers.  Newly launched or soon-to-be launched space-based telescopes are tailored to data-collection challenges associated with specific scientific goals. These instruments provide massive new surveys resulting in new catalogues containing terabytes of data, high resolution spectrograph and imaging across the electromagnetic spectrum, and incredibly detailed movies of dynamic and explosive processes in the solar atmosphere. These new data streams are helping scientists make impressive strides in our understanding of the physical universe, but at the same time are generating massive data-analytic and data-mining challenges for scientists who study them.

The Data Science Institute is working with researchers from the Imperial SpaceLab, and the Imperial Centre for Inference and Cosmology (ICIC), on utilising new technological advances in data science for their work.

Check out our case study on:

Smart Cities

Smart Cities encompass approaches to increasing efficiency in an urban infrastructure, where the efficiency gains are sought through the intelligent management, use of computing technology, and citizen participation, much of which is based around utilising urban data. The drive toward Smart Cities alongside the rising adoption of pervasive sensors is leading to big sensor data, which is so large and complex that it becomes difficult to use traditional methods to utilise it. For a modern city, a torrent of data is collected from sensors in various domains everyday. Sensors are now the dominant source of worldwide-generated data, with 1,250 billion gigabytes (1.25M petabytes) in 2010. And it’s growing exponentially. Sensor data has also high-velocity (collected and processed in real-time) and high-variety (collected by networks of diverse kinds of sensors). As these sensor datasets are interconnected with each other, the volume of the integrated data becomes even larger.

For example, at the Data Science Institute we explore these key challenges and provide a response through a sensor data software platform called Concinnity. Concinnity takes sensor data from collection to final product via a cloud-based data repository and an easy-to-use workflow designer. It supports rapid development of applications built on sensor data using data fusion and the integration of models to form novel workflows. These features enable value to be quickly derived from sensor data to drive the development of digital services in a smart city. Concinnity is key to the infrastructure used in the UK Digital Economy project, Digital City Exchange.

Check out our case studies on: