Inferring drivers-to-target interactions from time-series data using random forests
Olivier Martin (IPS2, Gif-sur-Yvette)
Although co-expression networks are often used to provide clues about genetic interactions, the study of longitudinal (time-dependent) data offers the possibility of extracting causal interactions from drivers to any candidate target. I will first present and benchmark our improved algorithm that builds on dynGENIE3 using random forests for regressions. I will then apply this approach to study the early stages of lateral root formation in Arabidopsis.