[ccp4bb]: How cal I get the missing part through refinement?

[Cheng-Yu Chen <b851615@life.nthu.edu.tw>]

***  For details on how to be removed from this list visit the  ***
***    CCP4 home page http://www.dl.ac.uk/CCP/CCP4/main.html    ***

Dear ccp4 users :

      I have a complete model and going to do the refinement using
Refmac. The complete model is obtained from arp_warp_5.0 autobuilding
mode and after that I use O to manual build the model. After I got the
complete model( it's a tetramer), I use arp_warp solvent building mode
to add water and the R was down to 0.17(the resolution is about 1.95). I

have run 50 solvent adding cycles and it's very wierd that I got 1600
waters in my model(it's a tetramer with a monomer about 300 residues).
      Then I checked the 2fofc map and the model from arp_warp. I found
that there are only part densities of some side chain and a missing
region of main chain of one monomer. So I try to refine the model using
Refmac 4.0(it's my first time to use it) and follow the examples in the
document. I use example 6 first to try to average the map of each
monomer and try to find the missing part. After trying several cycles,
the R was converg to 0.23 and the I got the averaged map of each
monomer. Based on the map, I can build the missing residues of some
part. But I want to refine the whole tetramer and not the NCS monomer so

I use example 5 to try to refine the whole tetramer. I try to add define

some parameters in the refinement and try several different cycles. But
after that the missing part was still there even the R value was down to

0.18 and Rfree was down to 0.24. I wondered that I did the wrong refine
procedures or I didn't set the correct keywords.
      I am not experienced with refinement. Could you tell me how to
find the missing part? (It could be a very general question). I have
read the document and I wondered that the NONX keyword can help me. But
I don't understand the document about the NONX keyword say. Could you
tell me more about it or some references? Another question is about the
water, after I run the example 5 of refmac with 50 cycles. I got more
than one thousand water molecules in my model. Is that normal or is that

another mistake I do in refinement?
      Thanks for your reading this long message. I'll appreciate any
suggestion and I think it will be very useful to me!! :)


Cheng-Yu Chen
Dept. of Life. Science
National Tsing Hua University, Taiwan, ROC.
email : b851615@life.nthu.edu.tw