PACS 05.40.a -Fluctuations phenomena, random processes, noise, and Brownian motion PACS 87.18.Tt -Noise in biological systems PACS 87.16.Nn -Motor proteins (myosin, kinesin dynein)Abstract. -We derive a new fluctuation-dissipation relation for non-equilibrium systems with long term memory. We show how this relation allows one to access new experimental information regarding active forces in living cells that cannot otherwise be accessed. For a silica bead attached to the wall of a living cell, we identify a cross-over time between thermally controlled fluctuations and those produced by the active forces. We show that the probe position is eventually slaved to the underlying random drive produced by the so-called active forces.Living cells are paradigmatic out of equilibrium systems, subjected to the ATP-driven activity of a collection of molecular motors, whose individual motion cannot easily be disentangled from thermal fluctuations. These are relatively small systems for which fluctuation phenomena are prominent, as investigated in recent works [1][2][3][4][5][6][7][8][9], with specific focus on the active forces and non-Newtonian rheology [1-3, 10] which can be measured via micro-rheological devices. Our goal in the present work is to show how the cell body's random pull-andpush can be investigated by means of very recent theoretical advances in the field of non-equilibrium statistical mechanics. And indeed, from the theory standpoint, in the recent past, much effort has been invested in deriving simple generalizations or extensions of the celebrated fluctuation-dissipation theorem for systems that can be arbitrarily far out of equilibrium [5,[11][12][13][14][15][16][17][18][19][20][21][22][23]. These efforts have given birth to a flurry of formulas relating the response of a system to a small external perturbation to some correlation functions, even when the system is not in an equilibrium state. Yet, so far none of these formulas has been of any predictive power in an actual experimental system. Existing experiments revolving around these theoretical advances have been confined to refined and nontrivial confirmations that in some small scale systems such as an optically trapped Brownian particle [24][25][26] the various ingredients entering these extended fluctuation-dissipation relations (EFDR) can indeed be measured. Given that the dynamics of (a) Correspondig Author: paolo.visco@univ-paris-diderot.fr living cells exhibit strong memory effects, we will first have to derive our own version of an EFDR adapted to a system with stochastic yet non-Markovian dynamics. The latter EFDR and its consequence for the understanding of active forces are the central topics of this letter. For instance, how relevant is the long term memory in relating response and fluctuations? What new pieces of information on non-equilibrium forces can be learnt from such a relationship? Can thermal fluctuations be disentangled from those arising from the active forces in a quantitative way? These are the questions we wish to address i...