Konrad Rawlik PhD

Konrad Rawlik


Publications:
2016
  Imputation of DNA Methylation Levels in the Brain Implicates a Risk Factor for Parkinson's Disease
Rawlik, K, Rowlatt, A & Tenesa, A 2016, 'Imputation of DNA Methylation Levels in the Brain Implicates a Risk Factor for Parkinson's Disease' Genetics, vol 204, no. 2, pp. 771-781. DOI: 10.1534/genetics.115.185967

Understanding how genetic variation affects intermediate phenotypes, like DNA methylation or gene expression, and how these in turn vary with complex human disease provides valuable insight into disease aetiology. However intermediate phenotypes are typically tissue and developmental stage specific, making relevant phenotype difficult to assay. Assembling large case-control cohorts, necessary to achieve sufficient statistical power to assess associations between complex traits and relevant intermediate phenotypes, has therefore remained challenging. Imputation of such intermediate phenotypes represents a practical alternative in this context. We used a mixed linear model to impute DNAm levels of four brain tissues at up to 1826 methylome wide sites in 6259 Parkinson's disease patients and 9452 controls from across five GWAS studies. Six sites, in two regions, were found to associate with Parkinson's disease for at least one tissue. While a majority of identified sites were within an established risk region for Parkinson's disease, suggesting a role of DNAm in mediating previously observed genetic effects at this locus, we also identify an association with four CpG sites in chromosome 16p11.2. Direct measures of DNAm in the Substantia Nigra of 39 cases and 13 control samples were used to independently replicate these four associations. Only the association at cg10917602 replicated with a concordant direction of effect (P=0.02). cg10917602 is and 87kb away from the closest reported GWAS hit. The employed imputation methodology implies that variation of DNAm levels at cg10917602 is predictive for Parkinson's disease risk, suggesting a possible causal role for methylation at this locus. More generally this study demonstrates the feasibility of identifying predictive epi-genetic markers of disease risk from readily available datasets.


General Information
Organisations: Royal (Dick) School of Veterinary Studies.
Authors: Rawlik, Konrad, Rowlatt, Amy & Tenesa, Albert.
Keywords: (DNA methylation, imputation, Parkinson’s disease. )
Pages: 771-781
Publication Date: Oct 2016
Publication Information
Category: Article
Journal: Genetics
Volume: 204
Issue number: 2
ISSN: 0016-6731
Original Language: English
DOIs: 10.1534/genetics.115.185967
2013
  Path Integral Control by Reproducing Kernel Hilbert Space Embedding
Rawlik, K, Toussaint, M & Vijayakumar, S 2013, Path Integral Control by Reproducing Kernel Hilbert Space Embedding. in International Joint Conference in Artificial Intelligence (IJCAI).
General Information
Organisations: Institute of Perception, Action and Behaviour .
Authors: Rawlik, Konrad, Toussaint, Marc & Vijayakumar, Sethu.
Publication Date: 2013
Publication Information
Category: Conference contribution
Original Language: English
2012
  On Stochastic Optimal Control and Reinforcement Learning by Approximate Inference
Rawlik, K, Toussaint, M & Vijayakumar, S 2012, On Stochastic Optimal Control and Reinforcement Learning by Approximate Inference. in Robotics: Science and Systems VIII (RSS 2012). pp. 1-8.
We present a reformulation of the stochastic optimal control problem in terms of KL divergence minimisation, not only providing a unifying perspective of previous approaches in this area, but also demonstrating that the formalism leads to novel practical approaches to the control problem. Specifically, a natural relaxation of the dual formulation gives rise to exact iterative solutions to the finite and infinite horizon stochastic optimal control problem, while direct application of Bayesian inference methods yields instances of risk sensitive control. We furthermore study corresponding formulations in the reinforcement learning setting and present model free algorithms for problems with both discrete and continuous state and action spaces. Evaluation of the proposed methods on the standard Gridworld and Cart-Pole benchmarks verifies the theoretical insights and shows that the proposed methods improve upon current approaches.
General Information
Organisations: Institute of Perception, Action and Behaviour .
Authors: Rawlik, Konrad, Toussaint, M. & Vijayakumar, S..
Number of pages: 8
Pages: 1-8
Publication Date: 2012
Publication Information
Category: Conference contribution
Original Language: English
2011
  Stiffness and temporal optimization in periodic movements: An optimal control approach
Nakanishi, J, Rawlik, K & Vijayakumar, S 2011, Stiffness and temporal optimization in periodic movements: An optimal control approach. in Intelligent Robots and Systems (IROS), 2011 IEEE/RSJ International Conference on. IEEE Computer Society, pp. 718-724. DOI: 10.1109/IROS.2011.6094799
We present a novel framework for stiffness and temporal optimization of periodic movements, with an emphasis on exploiting the intrinsic passive dynamics to realize efficient actuation and control. We use a dynamical systems based representation tuned to the requirements of rhythmic movements and propose a systematic methodology to optimize for control commands, temporal aspect of movements and timevarying stiffness profiles from first principles of optimality. Evaluations on a single pendulum and underactuated two-link robot simulation highlight the benefits, achieving remarkable actuation efficiency on complicated, highly dynamic tasks such as swing-up and brachiation.
General Information
Organisations: Neuroinformatics DTC.
Authors: Nakanishi, J., Rawlik, Konrad & Vijayakumar, S..
Number of pages: 7
Pages: 718-724
Publication Date: 2011
Publication Information
Category: Conference contribution
Original Language: English
DOIs: 10.1109/IROS.2011.6094799
2010
  An Approximate Inference Approach to Temporal Optimization in Optimal Control
Rawlik, K, Toussaint, M & Vijayakumar, S 2010, An Approximate Inference Approach to Temporal Optimization in Optimal Control. in Proc. Advances in Neural Information Processing Systems (NIPS 2010). pp. 1-9.
Algorithms based on iterative local approximations present a practical approach to optimal control in robotic systems. However, they generally require the temporal parameters (for e.g. the movement duration or the time point of reaching an intermediate goal) to be specified a priori. Here, we present a methodology that is capable of jointly optimizing the temporal parameters in addition to the control command profiles. The presented approach is based on a Bayesian canonical time formulation of the optimal control problem, with the temporal mapping from canonical to real time parametrised by an additional control variable. An approximate EM algorithm is derived that efficiently optimizes both the movement duration and control commands offering, for the first time, a practical approach to tackling generic via point problems in a systematic way under the optimal control framework. The proposed approach, which is applicable to plants with non-linear dynamics as well as arbitrary state dependent and quadratic control costs, is evaluated on realistic simulations of a redundant robotic plant.
General Information
Organisations: Institute of Perception, Action and Behaviour .
Authors: Rawlik, Konrad, Toussaint, Marc & Vijayakumar, Sethu.
Number of pages: 9
Pages: 1-9
Publication Date: 2010
Publication Information
Category: Conference contribution
Original Language: English

Projects:
Impedance in the Optimal Feedback Control Framework of Human Motor Control (PhD)