Statistics and Data Science Hasselt University

Overview

Four complementary specializations are offered: Biostatistics, Bioinformatics, Quantitative Epidemiology and Data Science. All specializations provide a solid basis of data science, but the first three put more emphasis on statistics. The Statistics and Data Science specialization still has a good statistics basis, but offers more courses on other aspects of data science (e.g. data visualization, programming and algorithms, …).

Biostatistics

The specialization 'Biostatistics' focuses on statistical methods that are important for many different applications in the life sciences, including clinical trials.

Statistics in general, and biostatistics in particular, rests on solid mathematical and probabilistic foundations. This is why in both the first and second year, foundational courses are offered, in a step-up design, with the lighter versions offered during the first year. At the same time, the field’s strong focus on the bio-sciences is supported by a broad introduction to medical and molecular biology at Hasselt University .

The practicing biostatistician needs to be equipped with important modeling tools, such as linear models (regression, analysis of variance, etc.), generalized linear models (logistic regression, Poisson regression, etc.), multivariate methods, longitudinal data analysis methods, Bayesian methodology, time-to-event analysis, and so on. Evidently, fluency in the use of statistical software is expected, which is why not only dedicated courses but also assignments and course work throughout many courses focus on the computational aspects. Further, specialized courses are offered in clinical trials, omics data, spatial statistics, infectious diseases epidemiology, microbial risk assessment, and so on.

Biostatisticians must be able to communicate with researchers from various fields, report results, and give effective presentations. Developing such skills is an integral part of the program.

Bioinformatics

Technological developments in molecular biology over the last few decades have improved the knowledge of molecular and cellular processes underlying e.g diseases and responses to treatments. “Omics”-oriented approaches (such as genomics, transcriptomics, microbiome or proteomics) consider many molecules of a given type collectively instead of one molecule at a time, generating a system-wide understanding. These technologies can nowadays even be applied at a single cell level. Data obtained with the help of “omics” technologies are usually very voluminous (yielding even millions of measurements per single biological sample or per cell in a sample), highly structured, and complex.

Analysis of such data is not trivial and has become a specialty of its own. Of course, good knowledge of  statistical methodology is required and training in this respect is offered in the first year of our program. Additionally, an introduction to medical and molecular biology is offered, together with a decent training in programming. The second year focuses on the methods specific for the analysis of genomic,proteomic and microbiome data obtained by using technologies like next-generation sequencing, mass spectrometry, etc. Methods for integrative analyses of different types of data are considered, too.

Bioinformatics is an interdisciplinary science. Statisticians working in this domain need to be able to communicate with researchers of various fields, report results, and give effective presentations. Developing such skills is an integral part of the program.

Quantitative Epidemiology

The specialization 'Quantitative Epidemiology' focuses on the design and analysis of epidemiological studies, including the mathematical modelling of infectious diseases.

The design of epidemiological studies and  intervention measures, and the collection and analysis of epidemiological data require appropriate expertise in statistical methodology in combination with knowledge of other scientific disciplines such as medical biology, computer sciences, data management, social sciences, etc.

Statistical methodology for epidemiology rests on solid mathematical and probabilistic foundations. This is why foundational courses are offered, in a step-up design, during the first year, supported by a broad introduction to medical and molecular biology, linear models (regression, analysis of variance, etc.), generalized linear models (logistic regression, Poisson regression, etc.), multivariate methods, longitudinal data, Bayesian methodology, so on. An introduction to epidemiology is also provided in the first year. During the second year, in addition to three foundational courses, specialized courses are offered in spatial epidemiology, digital epidemiology, mathematical modelling of infectious diseases, environmental epidemiology and microbial risk assessment.

Evidently, fluency in the use of statistical software is expected, which is why not only dedicated courses but also assignments and course work throughout the courses focus on the computational aspects.

Statisticians must be able to communicate with researchers of various fields, report results, and give effective presentations. Developing such skills is an integral part of the program.

Data science

The specialization 'Data Science' is built on the handling, managing, visualizing and analysing many different types of complex and/or big data sources, with a focus on modern programming and computing environments, and with a solid knowledge of statistical principles.

With the advent of the big data era, several global challenges that were outside of reach can now start to be addressed. In the field of medicine, wearable devices and real-time sensors generate huge amounts of data that can shed light on triggers for disease episodes. Omics and genome sequencing can aid in managing and preventing diseases, especially if they are combined with other data sources such as information from social networks. Integrated analysis of weather data, credit card transactions and air pollution data sheds light on how people change their behaviour due to air pollution. Graph analysis of social network data makes it possible to identify fake accounts and fake news - a growing problem in the current political climate. The list goes on... A data scientist is someone who, apart from technical skills to tackle these issues, has a desire to dig deeper and go beneath the surface of a problem.

The Data Science specialization of the Master of Statistics and Data Science provides a comprehensive education in this field, covering the whole data science cycle from data gathering, cleaning and management, to analysis and visualisation, and finally dissemination. Apart from a very decent knowledge of statistical principles, the topics in the master therefore include (but are not limited to) data and software carpentry, programming in Python and R, statistics, algorithms, machine learning (including deep learning), and data visualisation. In addition to regular courses, students can integrate their knowledge and skills in several data science projects and a hack week.

Statisticians/data scientists must be able to communicate with researchers of various fields, report results, and give effective presentations. Developing such skills is an integral part of the program.

Key facts

The Master of Statistics and Data Science at Hasselt University is quite unique in the sense that (1) it offers a statistics education with a good sense of general data science, and that (2) it offers a data science specialization with a very sound understanding of important statistical concepts and solutions.