


... 


Panagiotis Papastamoulis 

Address:
Faculty of Biology, Medicine and Health 

Division of Informatics, Imaging and Data Sciences 

University of Manchester 

Oxford Road, Manchester M13 9PL, UK. 

Office:
B.1082 Michael Smith building. 

Education
2003: 

Dipl. in
Mathematics, Department of Mathematics, University
of Patras, Greece 
2005:


MSc in Applied
Statistics, Department of Statistics and Insurance Science, University
of Piraeus, Greece 
2010:


Ph.D. in
Statistics, Department of Statistics and Insurance Science, University
of Piraeus, Greece 
Academic Positions
2011  2012: 

Research Associate at URGV  Plant Genomics Research, INRA, Evry, France. 
2012  : 

Research Associate at the University of Manchester, FBMH: informatics, imaging and data sciences. 
Research Interests
I am interested in mixture model theory and estimation both from a Bayesian and a frequentist perspective.
My PhD thesis proposed a solution to the label switching problem in Bayesian analysis of mixture models as
well as a modification of the reversible jump MCMC algorithm for univariate normal mixtures.
As a post doc researcher at URGV plant genomic unit I developed an initialization scheme of the EM algorithm
for the efficient estimation of Poisson GLM mixtures. The method has been applied to real high throughput
sequencing data with large number of components.
Currently I'm working at the University of Manchester/Faculty of Life Science with Professor Magnus Rattray. The aim of our
project is the development of Bayesian methods for estimating transcript expression and performing differential expression analysis
in next generation sequencing data.
Publications in refereed international journals
1. Papastamoulis P. and Iliopoulos G. (2009). Reversible Jump MCMC in mixtures of normal distributions with the same component means. Computational Statistics and Data Analysis, 53: 900911.
2. Papastamoulis P. and Iliopoulos G. (2010). An artificial allocations based solution to the label switching problem in Bayesian analysis of mixtures of distributions. Journal of Computational and Graphical Statistics, 19: 313331.
3. Papastamoulis P. and Iliopoulos G. (2013). On the convergence rate of Random Permutation Sampler and ECR algorithm in missing data models.
Methodology and Computing in Applied Probability, 15(2): 293304.
4. Papastamoulis P. (2014). Handling the label switching problem in latent class models via the ECR algorithm.
Communications in Statistics, Simulation and Computation, 43(4): 913927.
5. Papastamoulis P., Hensman, J., Glaus, P. and Rattray, M. (2014). Improved Variational Bayes inference for transcript expression estimation. Statistical Applications in Genetics and Molecular Biology, 13(2): 203216.
6. *Hensman, J., *Papastamoulis P., Glaus P., Honkela A. and Rattray M. (2015). Fast and accurate approximate inference of transcript expression from RNAseq data. Bioinformatics, 31(24): 38813889
7. Papastamoulis P., MartinMagniette, M.L. and MaugisRabusseau, C. (2016). On the estimation of mixtures of Poisson regression models with large number of components. Computational Statistics and Data Analysis, 93 (3rd Special Issue on Advances in Mixture Models): 97106.
8. Papastamoulis P. (2016). label.switching: An R Package for Dealing with the Label Switching Problem in MCMC Outputs. Journal of Statistical Software, 69(1): 124.
9. Papastamoulis P. and Rattray M. (2017). A Bayesian model selection approach for identifying differentially expressed transcripts from RNASeq data. Journal of the Royal Statistical Society, Series C, doi: 10.1111/rssc.12213. preprint: arXiv:1412.3050 [stat.ME]
10. Papastamoulis P. and Rattray M. (2017). BayesBinMix: an R Package for Model Based Clustering of Multivariate Binary Data. The R Journal, 9(1): 403420. R script for reproducing the results.
Working Papers
11. Papastamoulis P. and Rattray M. (2017). Bayesian estimation of Differential Transcript Usage from RNAseq data. (submitted) arXiv:1701.03095 [qbio.GN]
12. Papastamoulis P. (2017). Overfitting Bayesian Mixtures of Factor Analyzers with an Unknown Number of Components. (submitted) arXiv:1701.04605 [stat.ME]
Miscellaneous updates  Upcoming events
Software
1. Papastamoulis P. and Iliopoulos G. (2010). ecr_urb.Rnw: Rweave code for applying the ECR algorithm to simulated MCMC output of univariate normal mixture models. Suplementary material of the article: An artificial allocations based solution to the label switching problem in Bayesian analysis of mixtures of distributions. Journal of Computational and Graphical Statistics, 19: 313331.
2. Papastamoulis P., MartinMagniette M.L and MaugisRabusseau C. (2012). poisson.glm.mix: R package for the estimation of high dimensional mixtures of Poisson GLMs.
3. Papastamoulis P., Hensman, J., Glaus, P. and Rattray, M. (2013). gen_dir_vb: C++ source code for approximating the posterior distribution of mixture weights using Variational Bayes.
4. Papastamoulis P. (2013). label.switching: R package for dealing with label switching problem in MCMC outputs of mixture models.
5. Papastamoulis P. and Rattray, M. (2014). rjBitSeq: Reversible Jump MCMC algorithm for Differential Expression Analysis
6. Papastamoulis P. (2016). BayesBinMix: Bayesian Estimation of Mixtures of Multivariate Bernoulli Distributions.
7. Papastamoulis P. (2017). fabMix: Overfitting Bayesian Mixtures of Factor Analyzers.