ÌÇÐÄTV

Previous Study

• PhD in Statistics & Complexity Science, University of ÌÇÐÄTV, 2007-2012.
  Title: Sparse Graphical Models for Cancer Signalling.
  Supervised by .
• Master of Science (MSc) in Complexity Science, University of ÌÇÐÄTV, 2007-2008.
  Graduated with Distinction.
  Details of my two mini research projects, undertaken as part of the MSc, can be found here.
• Master of Mathematics (MMath), University of ÌÇÐÄTV, 2003-2007.
  Graduated with 1st class honours.

Research Interests

The application of statistical and computational methods to probe interesting questions in molecular cancer biology. This typically involves analysing noisy, high-dimensional, small sample size data.

Key topics of interest are:

• Bayesian networks, network inference and graph structure search
• Bayesian variable selection
• Clustering and subtype discovery
• Informative prior distributions

Publications

• 
  S. M. Hill, , Y. Lu, J. Molina, L. M. Heiser, P. T. Spellman, T. P. Speed, J. W. Gray, G. B. Mills and S. Mukherjee (2012)
  Bioinformatics 28:2804-2810 []
  
• Sparse Graphical Models for Cancer Signalling
  S. M. Hill (2012)
  PhD Thesis. Centre for Complexity Science and Department of Statistics.
  University of ÌÇÐÄTV.
•   
  S. M. Hill, R. M. Neve, N. Bayani, W-L. Kuo, S. Ziyad, P. T. Spellman, J. W. Gray and S. Mukherjee (2012)
  BMC Bioinformatics 13:94 []
•  
  S. Mukherjee & S. M. Hill (2011)
  Bioinformatics 27:994-1000 []
• [pre-print]
  D. Barker, S. M. Hill and S. Mukherjee (2010)
  In Proceedings of the 5th IAPR international Conference on Pattern Recognition in Bioinformatics.
  Tjeerd M.H. Dijkstra, Evgeni Tsivtsivadze, Elena Marchiori and Tom Heskes (eds.)
  Lecture Notes in Bioinformatics 6282, Springer
• (book chapter)
  S. Mukherjee, T. P Speed and S. M. Hill (2010)
  In: Medical biostatistics for complex diseases. Emmert-Streib, F. and Dehmer, M. (eds.)
  Weinheim, Germany: Wiley VCH. pp347-372.
• (open access)
  C. Grabow, S. M. Hill, S. Grosskinsky, and M. Timme (2010)
  Europhysics Letters 90:48002.
  (This publication contains work from my first MSc Mini-project).

In Preparation

• Network-based clustering with mixtures of L1-penalized Gaussian graphical models: an empirical investigation
  S. M. Hill and S. Mukherjee
  (in prep.)
  

---

Selected conference contributions

• Data-driven characterisation of protein signalling networks in cancer
  S. M. Hill, Y. Lu, J. Molina, G. B. Mills and S. Mukherjee
  Netherlands Bioinformatics Conference, 2012 ()
  Lunteren, The Netherlands, 24-25 April 2012
  Poster presentation
• An exact empirical Bayes approach for incorporating biological knowledge into network inference
  S. M. Hill and S. Mukherjee
  Machine Learning in Systems Biology, 2011 (MLSB2011)
  Vienna, 20th - 21st July 2011
  Talk (proceedings)

• Inferring dynamic protein signalling networks in individual cancers
  S. M. Hill, Y. Lu, G. B. Mills and S. Mukherjee
  NCI ICBP Junior Investigator Meeting 2010
  Houston, TX, 29^th September – 1^st October 2010
  Speed-talk

• Dynamic Bayesian analysis of protein signalling connectivity reveals heterogeneity within single breast cancer subtype
  S. M. Hill, Y. Lu, J. Molina, G. B. Mills and S. Mukherjee
  Proceedings of the European Conference on Complex Systems, 2010 ()
  Lisbon University Institute, 13^th - 17^th September 2010
  Poster presentation (Track E - main conference)
  Talk at conference satellite:

• Dynamic Bayesian analysis of protein signalling connectivity reveals heterogeneity within single breast cancer subtype
  S. M. Hill, Y. Lu, G. B. Mills and S. Mukherjee
  Proceedings of the Leeds Annual Statistical Retreat, 2010 ()
  University of Leeds, 6^th - 8^th July 2010
  Poster presentation
• Inferring protein signalling networks in cancer using dynamic Bayesian networks
  S. M. Hill, Y. Lu, G. B. Mills and S. Mukherjee
  Proceedings of the European Conference on Complex Systems, 2009 (ECCS09)
  University of ÌÇÐÄTV, 21^st - 25^th September 2009
  Poster presentation 

---

Invited Talks

• Dynamic Bayesian analysis reveals protein signalling connectivity in individual cancers.
  Biomedical and Biological Systems Laboratory, School of Engineering, University of ÌÇÐÄTV.
  December 2010

• Dynamic Bayesian analysis reveals protein signalling connectivity in individual cancers.
  Complexity Forum, Centre for Complexity Science, University of ÌÇÐÄTV.
  November 2010

• Complexity and Statistical Analysis: Network and Combinatorial Inference.
  Healthcare and Complexity Workshop, University of ÌÇÐÄTV.
  December 2008

 

Contact: