I am currently involved with simulating observations by both ALMA and the SKA. Eventually some results will go here, but in the meantime I'll be posting miscellaneous bits that might be interesting to other people interested in simulating observations.
My summer student, Charles Romero, and I are switching from miriad to CASA partly because of CASA's python interface. That encourages scripting to automate as much as possible:
The FITS files written by IDL often don't include header information that CASA expects. This function (based on what Kumar Golap suggested) tries to get an IDL 'fits' file into casa so it can be used as an input image for almasimmos. It is a work in progress, and you may need to massage its input using miriad and/or karma (including kvis).
I snipped this out of almasimmos to make and CLEAN an image from a measurement set.