In molecular science and other disciplines vast amounts of data building a times series can be produced but often not be interpreted due to its size. Markov state models (MSM) are coarse graining strategies allowing to describe the time series with less degrees of freedom, such that relevant features can be extracted. Often these methods reiy on the assumption, that the underlying time series is a reversible Markov chains, since this assumption allows for a projection of the data onto a low dimensional space.

This assumption is crucial and no always satisfied like eye tracking data or non equilibrated biological systems. Within this project we developed GenPCCA allowing a coarse grained description without the reversibility assumption on the

data.