[phenixbb] real space rotamer restrains

Leonid Sazanov sazanov at ist.ac.at
Fri Jan 22 09:43:34 PST 2021


Dear Pavel,

OK, will send you files. Normally we use older phenix version, so some 
parameters for this run with the latest version may be off.

But would be still nice to be able to understand the meaning of all 
parameters.

Best

Leonid


On 22.01.21 18:30, Pavel Afonine wrote:
> Hi Leonid,
>
> with the latest version everything should just work at least the same 
> as before and at most much better compared to the previous version. If 
> this is not what you observe -- something must have gone wrong. Please 
> send me input files (model before and after refinement, map and map 
> resolution) and I will investigate as soon as I get files.
>
> Pavel
>
> On 1/22/21 08:58, Leonid Sazanov wrote:
>> Hi,
>>
>> Is there somewhere a description of all the new parameters in 
>> phenix.real_space_refine rotamer restrains?
>>
>> I mean
>>
>> rotamers {
>>     fit = all *outliers_or_poormap outliers_and_poormap outliers poormap
>>     restraints {
>>       enabled = True
>>       sigma = Auto
>>       target = max_distant min_distant exact_match fix_outliers
>>     }
>>     tuneup = outliers outliers_and_poormap
>>   }
>>
>>
>> What is poormap?
>>
>> What is the range of Sigma?
>>
>> What are those types of target?
>>
>>
>> With the latest nightly build, with default parameters starting model 
>> with 1% of rotamer outliers ends up with 10% (not using local_search, 
>> as it was done before).
>>
>> To minimise number of outliers, which parameters can be changed?
>>
>> Lower Sigma? Switch on tuneup?
>>
>>
>> Thanks for any info!
>>
>> Leonid
>>
>>
>
-- 
Prof. Leonid Sazanov FRS
IST Austria
Am Campus 1
A-3400 Klosterneuburg
Austria

Phone: +43 2243 9000 3026
E-mail: sazanov at ist.ac.at



More information about the phenixbb mailing list