Add to ORKGDirect numerical simulation (DNS) and large-eddy simulation (LES) of particle-laden turbulentchannel flow, in which the particles experience a drag force, are performed. In this flowturbophoresis leads to an accumulation of particles near the walls. It is shown that the turbophoresisin LES is reduced, in case the subgrid effects in the particle equations of motion are ignored. Toalleviate this problem an inverse filtering model is proposed and incorporated into the particleequations. The model is shown to enhance the turbophoresis in actual LES, such that a goodagreement with the DNS prediction is obtained