Development of new computational approaches based on artificial evolution to exploit massively parallel data obtained by proteomic analysis

In the LSMIS, we develop new mass spectrometry approaches to identify new  potential drug targets (proteins) including comparisons of proteomes from cells pathogenic and non-pathogenic. In this context, we have developed a protocol based on a multienzyme in-gel digestion coupled with nanoLC-MS-MS analysis to access to the complete characterization of the studied proteins. For the main proteins studied, such improvements should allow to access to more than 90% of their amino acid sequence and their post-translational modifications (glycosylation, lipidations) playing a major role in their activity or in their interactions modulation with different partners. However, the currently available algorithms are limited to interpret this new type of experimental data, because primarily using large databases in which the searched sequences have to be founded, which is a problem if the sequences are sought therein not. Our goal is therefore now to develop a new algorithm suitable for the interpretation of these new data to obtain a complete characterization of the proteins without the use of proteins databases (as is the case for point mutations occurring in each individual).

The algorithm proposed in this project will be based on massively parallel artificial evolution. It will be developed by the BFO team of the ICUBE laboratory (University of Strasbourg) to find plausible sequences compatible with the experimental masses.

Contact