Re: [ccp4bb]: DM - NCS averaging

[Kevin Cowtan <cowtan@ysbl.york.ac.uk>]

***  For details on how to be removed from this list visit the  ***
***          CCP4 home page http://www.ccp4.ac.uk         ***

On Wednesday 07 November 2001  4:05 pm, Ricardo Aparicio wrote:
>   0.998 0.425
>   0.439 0.998

Looks good. No problem with your mask or ops.

> I've found that when I use  MODE HIST MULT AVER, correlations assume
> oscillating and negative values.
> For exactly the same script but changing the mode to MODE HIST MULT AVER
> SOLV, everything is OK.

MULT is no longer recommended. Turn it off. (It became redundent with the 
perturbation gamma correction).

The behaviour of your ncs-correlation is extremely unusual. If the same thing 
happens without MULT, I'd be interested to see the complete logfile.

> 2) It seems to be wise to take advantage of NCS BEFORE doing solvent
> flattening, in other words, to make dozens of AVER cycles before use of
> solvent flattening option. Is this correct?

No.

'dm' will actually do averaging without solvent flattening just fine. The 
ncs-related regions will be averaged, and the remaining regions will be 
untouched.

But you probably don't want to do that - tests show that it is pretty 
ineffectual.

BTW, I actually recommend solve/resolve. It's much better than dm.