Seminars

Thursday, 28th January 2010

4pm - 5pm, Statistics Seminar Room 222, Science Centre

A. Prof. Marc Suchard - UCLA

Title: "Many-core statistical inference of stochastic processes: a bright computational future"

Abstract:Massive numerical integration plagues the statistical inference of partially observed stochastic processes. An important biological example entertains partially observed continuous-time Markov chains (CTMCs) to model molecular sequence evolution. Joint inference of phylogenetic trees and codon-based substitution models of sequence evolution remains computationally impractical. Parallelizing data likelihood calculations is an obvious strategy; however, across a cluster-computer, this scales with the total number of processing cores, incurring considerable cost to achieve reasonable run-time.

To solve this problem, I describe many-core computing algorithms that harness inexpensive graphics processing units (GPUs) for calculation of the likelihood under CTMC models of evolution. High-end GPUs containing hundreds of cores and are low-cost. These novel algorithms are particularly efficient for large state-spaces, including codon models, and large data sets, such as full genome alignments where we demonstrate up to 150-fold speed-up. I conclude with a discussion of the future of many-core computing in statistics and touch upon recent experiences with massively large and high-dimensional mixture models.



News

26th January 2010

4th Asia-Pacific conference on nutrigenomics: Genes-diet and gut health, Auckland, 21-24 February, 2010

Programme, registration forms and more information on speakers is available at www.nutrigenomics.org.nz

Gut health is a major and complex health problem in developed and developing nations, that involves microbes, genetic susceptibility and dietary intolerances. Nutrigenomics explains how food interacts with genes and ultimately provides individuals with information for better health. It’s a rapidly growing scientific discipline that can help the food industry develop and market high-value premium functional foods that match peoples’ genetic make-up.

The conference is an opportunity for the scientific community, health professionals, including gastroenterologists, pediatricians and dietitians to hear international researchers working at the cutting edge field of nutritional genomics (nutrigenomics).

Speakers include:

  • Prof Chris Mathew, Kings College, London, UK, will talk on the genetic basis of ulcerative colitis.
  • Prof Ian Rowlands, University of Reading, UK, will consider the effect of the microbiota and potential use of probiotics on colon inflammation and colon cancer.
  • Prof Jerry Wells, The University of Wageningen, The Netherlands, will focus on intestinal homeostasis and the role of microbe-host signalling in the epithelium.

The questions raised in this symposium are also of potential importance to the food industry, since gut health conditions, including Irritable bowel syndrome and Inflammatory bowel diseases are increasing in the western world, and substantially limiting available food choices. The programme on Wednesday the 24th will consider the question as to whether the marketplace is ready for nutrigenomic foods, with plenary talks from Jim Kaput, FDA, USA, Ben van Ommen, TNO, The Netherlands and Michael Fenech, CSIRO, Australia. It will also feature representatives from Nestle (Martin Kussman), Fonterra (Jeremy Hill), Alltech (Karl Dawson) and Danone, as well as from FRST, MoRST and FSANZ, finishing with a panel discussion on whether the marketplace is ready for nutrigenomic foods.


4th August 2009

Bioinformatics Summer Studentships - Closed!


The Bioinformatics Institute is once again offering summer studentships to undergraduate students who are enrolled in a BSc (Bioinformatics) and BSc (Hons). The scholarships provide funding over the summer break and a chance to experience real applications of Bioinformatics data. Students are invited to submit an academic CV (non-official) and to nominate the project they would be interested in undertaking (email Emma Marks at e.marks@auckland.ac.nz). Applications close on the 4th of Sept at 5pm. Notification of acceptance will be made via email in late September.

List of summer pojects can be download here Summer Studentship Projects 2009


4th August 2009

Here are some recent publications from members of the Bioinformatics Institute!

- Abbott, W. G. H.,Tsai, P.,Leung, E., Trevarton, A., ‘Ofanoa, M., Hornell, J., Gane, E. J., Munn, S.R., andRodrigo, A. G.(In press) Associations between HLA Class I Alleles and Escape Mutations in the Hepatitis B Virus Core Gene in New Zealand-Resident Tongans. Journal of Virology

- Rambaut, A; Ho, SYW;Drummond, AJ.Accommodating the Effect of Ancient DNA Damage on Inferences of Demographic Histories. MOLECULAR BIOLOGY AND EVOLUTION, 26 (2): 245-248 FEB 2009. Full text

- Atkinson, QD; Gray, RD;Drummond, AJ.Bayesian coalescent inference of major human mitochondrial DNA haplogroup expansions in Africa. PROCEEDINGS OF THE ROYAL SOCIETY B-BIOLOGICAL SCIENCES, 276 (1655): 367-373 JAN 22 2009. Full text

- Gray, RD;Drummond, AJ;Greenhill, SJ. Language Phylogenies Reveal Expansion Pulses and Pauses in Pacific Settlement. SCIENCE, 323 (5913): 479-483 JAN 23 2009. Full text

-Marks E. J., Rodrigo A. G.& Brunton D. H (In press). Ecstatic Display Calls of the Adélie penguin honestly predict male condition and breeding success. Behaviour.

-Wu, S.H.,Black, M.A., North, R. A., Atkinson, K.,R. andRodrigo, A. G.(In press). A statistical model to identify differentially expressed proteins in 2D PAGE gels. PLoS Comp. Biol.

- Marion Blumenstein1, Michael T. McMaster, Michael A. Black,Steven Wu,Roneel Prakash, Janine Cooney, Lesley M. E. McCowan, Garth J. S. Cooper and Robyn A. North. A proteomic approach identifies early pregnancy biomarkers for preeclampsia: Novel linkages between a predisposition to preeclampsia and cardiovascular disease. Proteomic

-Hayward, J. H.andRodrigo, A. G.(In press). Molecular epidemiology of feline immunodeficiency virus in the domestic cat (Felis catus). Veterinary Immunology and Immunopathology

-Rodrigo, A. G.(2009) The coalescent. In The Phylogenetics Handbook 2nd Edition (eds. Salemi, M., Vandamme A.M, and Lemey, P). Cambridge University Press.

-Langhoff, P.,Authier, A., Buckley, T. R., Dugdale, J. S.,Rodrigo, A., and Newcomb, R.D. (In press). DNA barcoding of the endemic New Zealand leafroller moth genera, Ctenopseustis and Planotortrix. Molecular Ecology Notes. Full text

-Rodrigo, A. G., Tsai, P.,andShearman, H.2009. On the Use of Bootstrapped Topologies in Coalescent-Based Bayesian McMc Inference: A Comparison of Estimation and Computational Efficiencies. Evolutionary Bioinformatics 5:97-105. Full text

- Gillman, L. N., D. J. Keeling,H. A. Ross,and S. D. Wright. (In press). Latitude, elevation and the tempo of molecular evolution in mammals. Proceedings of the Royal Society B-Biological Sciences.

- Wright, S. D., L. N. Gillman,H. A. Ross,and D. J. Keeling. (In press). Slower tempo of microevolution in island birds: implications for conservation biology. Evolution.

- Anderson, M. G.,H. A. Ross,D. H. Brunton, and M. E. Hauber. (In press). Begging call matching between a specialist brood parasite and its host: A comparative approach to detect co-evolution. Biological Journal of the Linnean Society.

 Featured Research
Louis RanjardLouis Ranjard

My research interests include phylogeny and evolution biology. I'm currently working on bird song evolution. Learned bird songs are well studied examples of culturally transmitted traits. Every young bird acquires its song(s) through an imitation process of the neighboring birds. With time, slight changes in songs can accumulate. Therefore, subunits isolated from songs can be considered as memes. The study of the distribution of these memes, in time and space, reflects information about the history of bird populations. The goal of this project is to apply computational and phylogenetic methods in order to investigate the songs evolution.

The project has two main components. First, specific bioacoustics and machine learning methods are developed for isolating memes from songs. Second, modern phylogenetic approaches are used to investigate the way songs evolve, for example, the evolution of songs can be modeled on trees. From these trees the rates of evolution can be inferred. We have been working on native species (New Zealand saddleback), and also well-studied foreign species (white-crowned sparrow) in order to validate our methods.
© 2006 The University of Auckland - all rights reserved