• Network Inference and Meta-analysis from expression data
In 2007, I invented the MRNet Network inference method, one of the first method i) able to infer a large-scale causal network but also ii) that uses a feature selection algorithm in order to build a network (a strategy that has become standard). In ten years, this seminal paper has reached 300+ citations. In 2008, we produced a related Bioconductor package that is now heavily downloaded and cited. In 2015, I supervised my PhD student Pau Bellot to produce three different conference papers focused on assembling networks in order to produce a meta-network. Two of these papers have been selected for becoming book chapters. In 2016 my PhD student Ngoc Cam Pham and I produced a paper on meta-analysis of expression data. The paper received a "best paper award" and got published in 2017 in the BMC BioDatamining journal.
  • Open-source Bioinformatics Tools
In 2008, I authored with G. Bontempi the MINet Bioconductor package, one of the first package to infer large-scale (up to several thousands) causal networks from Data. Ten years later, the package reaches between 5000 and 10000 downloads a year and its related papers get more than 30 new citations per year (an unusual success for our field). In 2009, I authored the infotheo R package, this package allows to compute information-theoretic measure in R and is among the top downloaded package of the CRAN ecosystem. In 2015, I supervised my first PhD student toward the Netbenchmark Bioconductor package, a tool that allows thorough and fast comparison of network inference algorithms. Two years later the related paper is in the top 25% of all research outputs scored by Altmetric for papers of the same age and source.
  • Systems Biology and Computational Biology

In 2010, I produced the Drosophila modENCODE transcriptional network, leading me to a co-first authorship of the related AAAS Science paper and the related 2012 Genome Research paper. Those two papers produce more than 100 citations per year. In 2017, my PhD student, Manuel Noll and I contributed to one of the first automatic, machine-learning-based plant-root annotation system that lead us to publications in Frontiers in Plant Science and in GIGAscience journals.

  • Feature Selection and machine learning

From 2005 to 2010, I authored several conference and journal papers on feature selection with G. Bontempi. We invented the DISR and MASSIVE feature selection methods, those two information-theoretic methods have been among the very first in the field to use explicitly the concept of complementarity (or synergy) of variables. This concept of complementarity/synergy of variables have since been widely used in feature selection leading our related papers to reach hundreds of citations in less than 10 years (which is unusual in this field).

  • Causality and Implications in Data Analysis
In 2010, I co-authored with G. Bontempi an ICML paper introducing the mIMR feature selection method that focuses on selecting causal variables. More recently, in 2014, I proposed the Rank Minrelation Coefficient, a new statistical measure specifically focused on selecting implicative variables. I have given several talk on this theoretical topic and to this day, my lab continue to explore the intricate connection between causality, implication, correlation and information.