(according to postings on the CCP4BB on Feb 5, 2010 by G. Sheldrick and P. Evans; slightly edited)
PROTEUM (and APEX) are the names Bruker uses to describe the instrument control software; they write out frames in Bruker format. In the standard Bruker system the frames are integrated by SAINT which outputs reflection records in Bruker .raw format (not to be confused with a different .raw format produced by Bruker powder systems). EVAL, d*TREK and possibly other integration programs can also process Bruker frames, but only SAINT and EVAL can currently integrate the diffraction patterns of non-merohedrally twinned and modulated structures (in up to 6 dimensions).
For scaling Bruker provides SADABS (for scaling normal and modulated structures) and TWINABS (for non-merohedral twins). SADABS provides a similar functionality to SCALEPACK and SCALA and outputs merged or unmerged reflection lists in various ASCII formats that can be converted to .mtz by e.g. the CCP4 program COMBAT. TWINABS writes SHELX HKLF 5 format .hkl files for refinement of non-merohedral twins with SHELXL and can also produce 'detwinned' HKLF 4 format files that can be input to CCP4 by COMBAT and then treated in the same way as data from untwinned crystals.. These are better quality than conventional 'detwinned' data because they use the SAINT partitioning of overlapping reflections, as a 'restraint' in the 'refinement' of the unique reflection intensities against the total intensities of the composite reflections (the 'observations'. In the group of G. Sheldrick, the resulting data have even been used successfully for Sulfur-SAD phasing.
From version 1.3.9, Pointless ( ftp://ftp.mrc-lmb.cam.ac.uk/pub/pre/ ) will convert the .raw files from SAINT to mtz for input into Scala (version 3.3.18 and later from the same FTP site).
It may be interesting to compare the data quality obtained via the SAINT-POINTLESS-SCALA and the SAINT-SADABS-(XPREP)-COMBAT routes.
XDS can process Bruker data.