Alain Schenkel

As of 1st of February 2008, I am no longer working in Liisa Holm's group.
I have moved to the Laboratory of Computational Engineering at EPFL, Switzerland.
My email address there is firstname.lastname@epfl.ch.

Ph.D., Researcher
Postdoctoral Fellow in Liisa Holm's Bioinformatics Group

Email: Firstname.Lastname@Helsinki.FI
Personal Homepage:
Telephone: (+358 9) 191 59117
Facsimile : (+358-9) 191 59366 (Departmental)

Mailing Address:
Institute of Biotechnology
Viikinkaari 5, P.O. Box 56
FIN-00014 University of Helsinki
Finland

Visiting Address:
Viikinkaari 5, Helsinki
Room D7-7009


My CV.

Scientific Activities: Publications

I work mainly on the protein folding problem. Being a necessary step towards the deciphering of proteins functions and interactions from genomic sequences, it is currently one of the most outstanding problem in biophysics. Traditionnally, two points of view have been considered for ab initio protein fold prediction: molecular dynamics and thermodynamical approaches. The former approach, consisting in numerically solving Newton equations, is currently out of reach in realistic situations due to lack of computing power, whereas the latter suffers from the difficulty of getting hold of realistic and accurate enough energy functions. One ingredient which has not been considered so far is the folding dynamics. Newly available experimental data are shedding light on this aspect and may help to unravel and isolate a few principles that drive the folding dynamics. Whether or not these recent progresses may lead to a significant improvement of current protein fold predictions is a particularly interesting question.

I am also in charge of the parallelisation of the GTG update procedure. GTG (Global Trace Graph) is an alignment graph obtained by the single-linkage clustering of all the residues of a representative set (40% identical) of all known proteins. Developed by L. Holm and A. Heger, it provides a novel transitive alignment algorithm based on consistency scores from pairwise alignments. Due to the ever-increasing number of known protein sequences, and as GTG is a very large data structure, its update currently requires high performance computing and involves the parallelization of the original GTG code. Computations are performed on the supercomputer IBM p690 at RZG, Germany, in the framework of the DEISA Extreme Computing Initiative.

In my daily work, I am programming mainly in C, C++, and Python (but still enjoying writing plain awk, bash scripts and such whenever I have the opportunity). My current interests in programming are object oriented and parallel programming.

See this page for my former activities in mathematical physics.


Back to the main page...