Computational Evolution Group

Two PhD scholarships are available in the Computational Evolution Group under the supervision of Dr Alexei Drummond,
co-founder of the BEAST software package for Bayesian evolutionary analysis.

The successful candidates will be involved in developing a Bayesian unification of geographical population dynamics and phylogenetics with a focus on human pathogenic viruses. Current phylogenetic models provide inferences about genetic variation, however this research will focus on the roles of landscape, geography, dispersal and migration, in shaping (viral) genetic histories. The successful candidates would work on two novel classes of tractable phylogenetic models that explicitly include the spatial distribution of samples and the geographic features of surrounding habitats. While the primary focus of the research will be on viral pathogens, the proposed work will have the potential to establish a new foundation for phylogeographic modeling in general. Besides developing methodology, the successful candidates will be involved in developing a new version of the BEAST software to include both phylogeographical modeling and visualization.

The research will be conducted in collaboration with Dr Marc Suchard’s research group in UCLA and the successful candidates will have the opportunity to spend one month a year in Los Angeles.

The successful candidates will have a background in applied mathematics, statistics, computer science or quantitative biology.

Students will be affiliated with either the Bioinformatics Institute or the Department of Computer Science depending on their background and preference.

Applications should be sent to Dr Alexei Drummond (alexei AT cs.auckland.ac.nz)
by 1st March 2009, and include a CV, and the names of two references.

Applications are invited for a one year full-time Masters scholarship in the area of statistical phylogenetics and genomics. The successful candidate will join the newly created Computational Evolution Group (David Bryant, Alexei Drummond and Stephane Guindon) at the University of Auckland, New Zealand.

The scholarship is funded by a Marsden Fund grant awarded to S. Guindon. The Masters program will start from 1 March 2009, consisting of 75% research and 25% coursework. The Masters research project consists of a phylogenomic approach to characterize the patterns of deviations from the molecular clock and the variations of selection regimes along genomes.

The scholarship is open to every student who has a Bachelors degree with Honours or equivalent in Bioinformatics, Biology, Statistics, Mathematics or Computer Science. The total amount of money is NZ$16,000 (tax free) plus domestic fees. An international student will need to pay the outstanding amount of international fees themselves. Each applicant should send her/his CV to the following address before 1 Febuary 2009, along with a cover letter that outlines her/his experience and the contact details of at least one referee. For more information, please contact:

Stephane Guindon

Department of Statistics

The University of Auckland

Private Bag 92019, Auckland

New Zealand

E-mail: guindon@stat.auckland.ac.nz

Phone: +64 9 3737599 ext. 82755

Carrasco et al (2007) report on the distribution of fitness effects of random point mutations in an RNA virus within its natural host environment. They demonstrates that most random point mutations in the Tobacco Etch virus are lethal or very deleterious (average fitness effect of non-lethal mutations is -0.4 (40%)), and confirm that, in RNA viruses, about 40% of point mutations are lethal in a wide variety of environments. Direct studies of the distribution of fitness effects are still very rare in the literature, despite their central importance in understanding the interplay between natural selection and stochastic effects on evolutionary dynamics. This study, along with a handful of previous studies begin to shed light on the essential features of the distribution of fitness effects in RNA viruses. The evidence points to RNA viruses being under the strongest purifying selection in the natural world, with almost half of random point mutations being lethal. In combination with Sanjuán, Moya, Elena (2004), this paper also provides quantitative support for the idea that the natural host environment is more selective than more artificial laboratory environments, such as culture media. This has implications for the design of experimental systems that aim to understand evolution in natural environment.

Alexei

The Computational Biology and Evolution group is advertising for a full-time research programmer. He/she should be proficient in Java and familiar with scientific programming, multi-threaded applications, XML and Swing GUIs. A knowledge of population genetics or phylogenetics would also be a plus. The programmer will support open source packages developed by members of the research group including BEAST. More details of the position can be found here

Cheers
Alexei

I recently came across the following web site: http://www.ncbi.nlm.nih.gov/books/bv.fcgi?call=bv.View..ShowTOC&rid=hmg.TOC&depth=10. This book may not be very up-to-date but it gives a great overview of the discipline. Thanks to the authors, Tom Strachan and Andrew P. Read, and the editors for agreeing to put these resources on line.

PhyML 3.0 is now available. We have spent more than one and a half year developing new algorithms to improve the tree search and assess internal branches. Our tests (submitted) indicate that this new version provides a significant gain, mainly in terms of likelihood scores, compared to the original version of the program.

Please feel free to check our new web server at: http://www.atgc-montpellier.fr/phyml and/or drop me an e-mail so that I can send you the sources.

Cheers,

-Stephane

This conference was great: excellent people to talk to and amazing venue.
The web site: http://www.lirmm.fr/MIEP08/

** This position has been filled. Thanks to all the applicants. **

An opportunity exists to work, as a Postdoctoral Research Fellow, on a project funded by an Australian Research Council Discovery grant. A motivated candidate with a PhD is required to undertake original research modelling the stochastic dynamics and evolutionary genetics of virus populations.

The research will take place jointly at Swinburne University of Technology and Melbourne University in Melbourne, Australia, in a new theoretical research group. These universities have a broad program of research, ranging from astrophysics to bioinformatics research, in conjunction with strong computational resources and a wide range of biological sciences institutions in Victoria.

This project will study theoretical population genetics and population dynamics from a viewpoint of modeling statistics and fluctuations. The project has a computational and quantitative focus, aided by close cooperation with experts in computational evolutionary biology at the University of Auckland. The candidate will be expected to travel to New Zealand to help the cooperative aspect of the research program.

The successful candidate will focus on the highly relevant case of modeling viral evolution in a single infected host. Theoretical methods will include Monte-Carlo and stochastic computational and analytic
techniques employing the Poisson representation as well as other methods. Familiarity with high-level computer languages would be useful, as would a knowledge of modern population genetics and the use of stochastic differential and Fokker-Planck equations.

You should possess a PhD in one of the following areas: population genetics, evolutionary biology, mathematics/mathematical biology, bioinformatics/computational biology, or statistical and computational physics with a willingness to learn the genetics background. The successful candidate should also have authored articles in international peer-reviewed journals.

This is a fixed term (two year), full-time appointment at Academic Level A/B, depending on qualifications and experience.

Applications close on June 1st, 2008.

Contact drummond@physics.uq.edu.au or alexei@cs.auckland.ac.nz

Peter D Drummond
Professor of Theoretical Physics,
ARC Professorial Fellow,
Scientific Director of the UQ node of the ARC Centre
of Excellence for Quantum-Atom Optics
University of Queensland
Australia

Alexei Drummond
Senior Lecturer in Bioinformatics
Department of Computer Science
University of Auckland
New Zealand

Last week I taught the Advanced Module of the inaugural Asia-Pacific Viral Phylogenetics & Bioinformatics Workshop hosted in the School of Public Health, Chinese University of Hong Kong. The course was organised into back-to-back Basic and Advanced Modules with 50 students in each module. The timetable was rather demanding for teachers and students alike, with the first day of each module involving 8 hours of lectures and the second day occupied by computer practicals in the lab.

Dr Julian Tang (Dept Microbiology, CUHK) organized the workshop with great efficiency and was very understanding when I left everything to the last minute. The other protagonist was Dr Stéphane Hué (Division of Infection and Immunity, University College London, London, UK) who taught the Basic Module with great success. He covered sequence alignment, basics of phylogenetics, bootstrapping, detection of recombination, and molecular epidemiology.

By the end of the first day of lecturing in the Advanced Module I had covered Maximum Likelihood and Bayesian phylogenetic inference, molecular detection of natural selection, molecular coalescent-based population genetics, measurably evolving populations, relaxed molecular clocks and viral phylodynamics. As the workshop banquet was the night before my lectures I started off at a measured pace, but by the late afternoon session I was back to my bouncing self and we finished with a flourish. When lecturing I tend to move around the room and I don’t like podiums. As a result I estimate that I walked about 15km in front of the audience that day, talking non-stop (well there were three breaks for good food and refreshment). Who says we don’t earn our pay in academia?

The second day of the Advanced module was in the lab and we covered PAUP*, MrBayes 3.1, Datamonkey/HYPHY and BEAST. The students were great, but there was a general consensus that although the course was very stimulating, we tried to do a bit too much. However, we got more than enough positive feedback to do it again, with a likely change in the format: lectures in the morning, labs in the afternoon (so as to consolidate the morning’s ideas). I am looking forward to it – especially the roast pigeon and double-boiled fish maw (which I like mainly for the name). Got to love that Chinese cuisine (although I said no to the shark fin soup – sharks’ get a bad enough rap as it is. It’s the pigeons we ought to be eating – they are numerous and tasty… hang on, I guess that is what the early American settlers said about the 5 billion passenger pigeons before they went extinct!)