AESOP : People / Dr David Thornley

Departmental Associate
DoC Home page

My research interests are essentially in numerical and probabilistic modeling of systems. A PhD investigating the behaviour of DNA sequencing data and the reactions behind it led me from numerical optimizations to process algebras and machine learning, and a post developing a queuing paradigm in performance modeling brought me up to scratch on probability manipulation. The people I've worked with have made my continuing development as a researcher thoroughly enjoyable. Peter Harrison has made a crucial contribution to my career, offering me a performance analysis position which has taken me to places I could not have predicted. Duncan Gillies has been a staunch collaborator in the DNA sequencing work, and currently employs me on a project looking at quality of information in sensor networks. The members of the Aesop research group have added enormously to the enjoyment of scientific endeavour.

18 Publications with AESOP

Journal Article

An approximate solution of PEPA models using component substitution: Nigel Thomas, Jeremy T. Bradley, David Thornley; IEE Proceedings - Computers and Digital Techniques; March, 2003

Conference and Workshop Papers

Machine learned regression for abductive DNA sequencing: David Thornley, Maxim Zverev, Stavros Petridis; The 2007 International Conference on Machine Learning and Applications; December, 2007
Novel anisotropic multidimensional convolutional filters for derivative estimation and reconstruction: David Thornley; Proceedings of The 2007 IEEE International Conference on Signal Processing and Communications (ICSPC 2007); November, 2007
Decoding Trace Peak Behaviour - A Neuro-Fuzzy Approach: David Thornley, Stavros Petridis; Proceedings of the IEEE International Conference on Fuzzy Systems; June, 2007
Exploring accuracy and correctness in solution to matrix polynomial equations in queues: David Thornley, Harf Zatschler; QEST 2006; August, 2006
Machine Learning in Basecalling - Decoding trace peak behaviour: David Thornley, Stavros Petridis; IEEE Symposium on Computational Intelligence in Bioinformatics and Computational Biology 2006; August, 2006
Modelling along the DNA template in the Sanger method: inhibition through competition and form: David Thornley; Process Algebra and Stochastically Timed Activities 2006; May, 2006
A novel approximated joint activity transition structure in a tandem feedback unreliable server queue: David Thornley, Harf Zatschler, Nigel Thomas; UKPEW'03; June, 2003
An automated formulation of queues with multiple geometric batch processes: David Thornley, Harf Zatschler, Peter G. Harrison; HETNETS'03; June, 2003
Analysis and enhancement of network solutions using geometrically batched traffic: David Thornley, Harf Zatschler; UKPEW'03; June, 2003
Synchronized negative customers in an unreliable server queue: David Thornley; HETNETS'03; June, 2003
Approximate solution of a class of queueing networks with breakdowns: Nigel Thomas, David Thornley, Harf Zatschler; European Simulation Multiconference, Nottingham; May, 2003
Geometrically batched networks: Peter G. Harrison, David Thornley, Harf Zatschler; ISCIS 2002; September, 2002
Queues with simultaneous loss on breakdown: David Thornley; Eighteenth Annual UK Performance Engineering Workshop, University of Glasgow; June, 2002

National Workshop Paper

An approximate solution of PEPA models using component substitution: Nigel Thomas, Jeremy T. Bradley, David Thornley; Eighteenth Annual UK Performance Engineering Workshop; June, 2002

PhD Thesis

Analysis of Trace Data from Fluoresence Based Sanger Sequencing: David Thornley; Department of Computing, Imperial College London; August, 1997

Department of Computing Technical Reports

Anisotropic Multidimensional Savitzky Golay kernels for Smoothing, Differentiation and Reconstruction: David Thornley; June, 2006
Trace modelling for abduction basecalling: David Thornley; June, 2006

Information from pubs.doc.ic.ac.uk/authors/djt.