So far I've only checked the phenix.refine output files (both mtz with map coefficients and calculated amplitudes). I ran into this with some of my own code automatically assuming the 5.0 compliance and using the HKL_base dataset to get the unit cell parameters. HTH, Ed. On Mon, 2010-08-16 at 16:15 -0700, Ralf W. Grosse-Kunstleve wrote:
Hi Ed,
MTZ-files generated by phenix (at least some of them) are missing the base dataset (HKL_base) which is a requirement since CCP4 5.0. Is this deliberate?
Could you be more specific? The iotbx.mtz library I wrote should create the HKL_base pseudo-dataset. But we also generate mtz files from solve, resolve, and phaser, which use the raw cmtzlib directly. If you know which phenix application didn't write the HKL_base it will save us some time reviewing the various mtz writers. Also, I'm curious to know where you run into difficulties with the old-style mtz files.
Ralf _______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb