[phenixbb] low resolution map weirdness with newer releases

28 Mar 2011


      Something changed between  build 432 and 486 that made my low resolution maps look much crappier. See attached tiff. See if you can guess which map was produced by the release 486.

def file looks more or less like this (from a --diff-params):

refinement {
 crystal_symmetry {
   unit_cell = top secret
   space_group = top secret
 }
 input {
   pdb {
     file_name = "../4_23_003.pdb"
   }
   xray_data {
     file_name = "../4_23.sca.mtz"
     labels = "4_23.sca,SIG4_23.sca"
     high_resolution = 3.9
     r_free_flags {
       file_name = "../4_23.sca.mtz"
       label = "R-free-flags"
       test_flag_value = 1
     }
   }
 }
 output {
   prefix = "test-486"
   serial = 1
 }
 refine {
   strategy = *individual_sites individual_sites_real_space rigid_body \
              individual_adp group_adp tls occupancies group_anomalous
   sites {
     individual = "chain A and resseq 100:107"
     individual = "chain B and resseq 500:507"
     individual = "chain E and resseq 100:107"
     individual = "chain F and resseq 500:507"
     individual = "chain C and resseq 111:117"
     individual = "chain D and resseq 111:117"
     individual = "chain G and resseq 111:117"
     individual = "chain H and resseq 111:117"
     individual = "chain C and resseq 63:68"
     individual = "chain D and resseq 63:68"
     individual = "chain G and resseq 63:68"
     individual = "chain H and resseq 63:68"
     rigid_body = "chain A"
     rigid_body = "chain B"
     rigid_body = "chain C"
     rigid_body = "chain D"
     rigid_body = "chain E"
     rigid_body = "chain F"
     rigid_body = "chain G"
     rigid_body = "chain H"
   }
 }
 main {
   number_of_macro_cycles = 1
   secondary_structure_restraints = True
   random_seed = 3141766
   optimize_mask = False
 }
 modify_start_model {
   adp {
     set_b_iso = 50
   }
 }
 fake_f_obs {
   mask {
     ignore_hydrogens = True
   }
 }
 geometry_restraints.edits {
   excessive_bond_distance_limit = 20
 }
 mask {
   ignore_hydrogens = True
 }
 simulated_annealing {
   start_temperature = 300
   final_temperature = 200
 }
 target_weights {
   wxc_scale = 0.4
 }
 ias {
   b_iso_max = 60
 }
 secondary_structure {
   h_bond_restraints {
     substitute_n_for_h = True
     remove_outliers = False
     sigma = 0.02
     h_o_distance_ideal = 2.95
     h_o_distance_max = 5
     n_o_distance_ideal = 2.95
     n_o_distance_max = 5
   }
}

[phenixbb] low resolution map weirdness with newer releases

Scott Classen