Also, see some discussion around Table 1 here: https://journals.iucr.org/d/issues/2013/04/00/dz5273/dz5273.pdf Different methods exist to split reciprocal space in resolutions shells: by volume, by shell width, by number of reflections and so on. Their utility depends on specific context. I'm not aware of specific guidances nor rules for defining the highest resolution bin. Pavel On 7/29/21 08:05, Johannes Cramer wrote:

Hi,

I think most programs just divide the number of reflections by 10 (or however many shells you want there to be) and make it so that all shells contain the same number of reflections.

Cheers, Johannes

Am Mi., 28. Juli 2021 um 21:10 Uhr schrieb Pavel Afonine mailto:[email protected]>:

Hi Smita,

>               Can someone suggest to me, how we can decide the > resolution range in the outer shell during the protein crystal data > set.

the choice is pretty arbitrary. Typically the software chooses it for you and it is fine most of the time.

Pavel

Hello Sir, It is a very nice webinar. Gives me the clearcut idea for statistics and how it will be useful. On Thu, Jul 29, 2021 at 11:04 PM Petr Kolenko wrote:

Dear Smita, Pavel is right. However, if you want do to it the optimal way, use paired refinement. Select conservative resolution first (e.g. cut your data at I/sigma equals 2). Build a reasonable model of your structure and run paired refinement afterwards using higher and higher resolution data. For that purpose, you may use program PAIREF (https://pairef.fjfi.cvut.cz/). PAIREF also works for PHENIX users and it is very easy to install. If you wanted to know more about the program in general, read here: https://journals.iucr.org/m/issues/2020/04/00/mf5044/index.html We had a webinar from which we made video available: https://pairef.fjfi.cvut.cz/dokuwiki/doku.php?id=webinar_2021-03 PHENIX contains paired refinement as well. It should be somewhere in the most advanced version of the refinement? It does the same thing, but the report from the job is focused differently. We report merging data statistics and some other parameters. Moreover, PHENIX itself does not allow complete cross-validation that we strongly recommend to the users. In any case, feel free to contact me directly. ;-) Best regards, Petr

________________________________________ From: [email protected] on behalf of Pavel Afonine Sent: Wednesday, July 28, 2021 9:10:36 PM To: Smita Yadav; [email protected] Subject: Re: [phenixbb] (no subject)

Hi Smita,

...

Can someone suggest to me, how we can decide the resolution range in the outer shell during the protein crystal data set.

the choice is pretty arbitrary. Typically the software chooses it for you and it is fine most of the time.

Pavel

_______________________________________________ phenixbb mailing list [email protected] http://phenix-online.org/mailman/listinfo/phenixbb Unsubscribe: [email protected]

-- With Regards Smita Yadav