Xavier Robin Email & Phone Number

Basel Area, Switzerland

As a scientific programmer in the SWISS-MODEL team, I participate in may aspects of the development, in particular around in ligands, and implemented an innovative visualization based on the NGL viewer.I am currently the lead developer of CAMEO, which provides a fully automated benchmarking service to protein modeling servers worldwide. Benchmarking has been a central tool in the community to improve the quality of protein structure models. I am currently focusing on the next major development in CAMEO: the benchmarking of ligand pose predictions within protein structures.My role also involves the generation and analysis of bioinformatics datasets, including data pre-processing, cleanup, statistical analysis and visualization.

Copenhagen Area, Denmark

Région De La Capitale Copenhague, Danemark

I am developing a Bayesian evidence network model to analyse phosphorylations and protein concentration. STRING and NetworKIN are used as prior information. I use an innovative Bayesian variance modelling of the uncertainties of the mass spectrometry ratios, followed with Monte Carlo simulations.

Copenhagen Metropolitan Area

As an independent postdoctoral researcher with a fellowship from the Swiss National Science Foundation, I worked with High Content Screening technologies (high-throughput microscopy) and advanced non-linear algorithms and bioinformatics programming for big data analysis. In particular, I implemented high dimension feature extraction from cell image data followed by an unsupervised Deep Belief Network (Deep Learning) for dimensionality reduction. The model was implemented in R and C++.

During this short post-doc, I completed the publication of the last tools and results obtained during my thesis.

During my thesis, I developed an algorithm to combine biomarkers with an approach based on thresholds. It was compared with established methods (SVM, logistic regression) with cross-validation and applied to several real-life datasets of brain diseases: stroke, subarachnoid hemorrhage and trypanosomiasis. I built a web interface allowing users without R programming skills to make use of its outstanding modelling capabilities, PanelomiX. I also performed data analysis and biostatitics in various other projects of the lab, related to quantitative analysis, statistics and mass spectrometry. I provided lectures to master's students in this topic.As part of this project, I developed pROC, an open-source package for R and S+ to analyse and compare ROC curves. I implemented bootstrapping and several published statistical methods. Development is ongoing with implementation sample size / power computations and various UI improvements. I took part in many projects of the lab, especially in data analysis. I implemented a published mass-spectrometry quantitation tool in the lab's data analysis workflow (Easyprot).

Geneva, Switzerland

I studied the applicability of biomarkers panels.I curated the stroke database that had been collected during the two previous years.

Internship in the Master of Proteomics and Bioinformatics. Creation of MIAPEGelDB, a web-based database for MIAPE documents with perl/Catalyst framework and PostgreSQL.http://miapegeldb.expasy.org/

Phd, Interdisciplinary, Science And Medicine

Activities and Societies: Webmaster of the ACERM associationFocus in bioinformatics and statistics.

Master, Proteomics And Bioinformatics

Bachelor, Biology

Maturité, Options Scientifiques