3 Year Funded PhD in Biomedical Informatics

The University of Edinburgh

Network-based approaches are increasingly relevant as a computational-biological paradigm for Precision Medicine, enabling the systematic incorporation of evidence from full-genome sequencing, genome-wide association studies, and other empirical sources into the investigation of disease, including potential drug and intervention targets1. This paradigm poses significant methodological and computational challenges. As more data become available, the analysis of such large-scale complex networks is hindered by limitations of existing statistical tools. While graph-theoretic and machine learning methods have been employed in a number of areas, new computational methods are needed which scale to meet the prediction and model-explanatory requirements of large, high-dimensional datasets.


This PhD will explore new methods that combine knowledge of network properties and Bayesian learning in order to tackle these challenges. In cooperation with other teams, these methods will be tested on different applications relevant to Precision Medicine, such as the stratification of patient-derived data for enabling clinically-oriented decisions and therapeutic research in cancer and neurodegenerative diseases. The student will:

  • Investigate the application of embedding methods which have been applied to large-scale natural language processing2,3 and more recently to the analysis of web data4, 5, 6 to molecular networks.
  • Devise inference methods based on these representations which improve upon existing methods for the identification of influential network nodes or edges, network modules, and other network-based properties of biological significance.
  • Investigate application cases in oncology and neurodegenerative disease research, which require the analysis of multiple sources of high-dimensional data, such as multi-omics datasets or different layers of information relevant to predicting patient outcomes.

Training outcomes

The successful candidate will have the opportunity to work in several areas:

  • Bioinformatics
  • Diverse “omics” datasets
  • Deep-learning methods
  • Cooperation with experts in oncology and neurodegenerative diseases
  • Graph theory and network science
  • Probabilistic graphical models


Dr  Luz (s.luz@ed.ac.uk), Dr Theodoratou (Institute of Genetics and Molecular Medicine), and Dr Azuaje (Luxembourg Institute of Health).


  1. Barabási, A.-L. et Al. Network medicine: a network-based approach to human disease. Nature Reviews Genetics 12,  (2011).
  2. Bengio Y., et al. Neural Probabilistic Language Models. In Innovations in Machine Learning.  (2016)
  3. Mikolov, T., et Al. Efficient Estimation of Word Representations in Vector Space. arXiv:1301.3781 (2013).
  4. Grover, A. & Leskovec, J. node2vec: Scalable feature learning for networks. ACM SIGKDD  855–864 (2016).
  5. Tang, J. et Al. Line: Large-scale information network embedding. Procs. WWW Conference, 1067–1077 (2015).
  6. Perozzi, B., et Al. Deepwalk: Online learning of social representations. ACM SIGKDD 701–710 (2014).


We invite applications from candidates with a good first degree (and ideally a Master's) in bioinformatics, computer science or related disciplines.

Applicants must meet the entry requirements  for acceptance to the University of Edinburgh PhD programme. See:


UK/EU tuition fees only (any eligible non-EU candidates must fund the remainder of the overseas tuition fee).


This studentship covers tuition fees at UK/EU rates (currently £4,195p.a.), additional costs, conference travel of up to £300p.a., and stipend at international rate (currently £14,553p.a.) for 3 years.

Application procedure

Please email:

  • a CV,
  • Personal statement indicating how you meet the criteria,
  • 2 academic references

to Dr Luz (s.luz@ed.ac.uk), CC s.georges@ed.ac.uk

Interviews will be held in Edinburgh (or by videoconference/Skype).
Start date: early 2018.
Closing date for applications: 11-Dec-2017

Share this PhD
  Share by Email   Print this job   More sharing options
We value your feedback on the quality of our adverts. If you have a comment to make about the overall quality of this advert, or its categorisation then please send us your feedback
Advert information

Type / Role: