Context. The Atacama Large Millimeter/submillimeter Array (ALMA) will have the necessary resolution to observe a planetary gap created by a Jupiter-mass planet in a protoplanetary disk. Because it will observe at submillimeter and millimeter wavelengths, grains in the size range 10 μm to 1 cm are relevant for the thermal emission. For the standard parameters of a T Tauri disk, most grains of this size range are weakly coupled to the gas (leading to vertical settling and radial migration) and the common approximation of well-mixed gas and dust does not hold. Aims. We provide predictions for ALMA observations of planet gaps that account for the specific spatial distribution of dust that results from consistent gas+dust dynamics. Methods. In a previous work, we ran full 3D, two-fluid Smoothed Particle Hydrodynamics (SPH) simulations of a planet embedded in a gas+dust T Tauri disk for different planet masses and grain sizes. In this work, the resulting dust distributions are passed to the Monte Carlo radiative transfer code MCFOST to construct synthetic images in the ALMA wavebands. We then use the ALMA simulator to produce images that include thermal and phase noise for a range of angular resolutions, wavelengths, and integration times, as well as for different inclinations, declinations and distances. We also produce images which assume that gas and dust are well mixed with a gas-to-dust ratio of 100 to compare with previous ALMA predictions, all made under this hypothesis. Results. Our findings clearly demonstrate the importance of correctly incorporating the dust dynamics. We show that the gap carved by a 1 MJ planet orbiting at 40 AU is visible with a much higher contrast than the well-mixed assumption would predict. In the case of a 5 MJ planet, we clearly see a deficit in dust emission in the inner disk, and point out the risk of interpreting the resulting image as that of a transition disk with an inner hole if observed in unfavorable conditions. Planet signatures are fainter in more distant disks but declination or inclination to the line-of-sight have little effect on ALMA’s ability to resolve the gaps. Conclusions. ALMA has the potential to see signposts of planets in disks of nearby star-forming regions. We present optimized observing parameters to detect them in the case of 1 and 5 MJ planets on 40 AU orbits.