Re: [ccp4bb] merging of datasets

[Kay Diederichs]

Dear Jordi,

for XDS/XSCALE users, there is XDSCC12 (see 
https://wiki.uni-konstanz.de/xds/index.php/Xdscc12 ).

It is integrated in XDSGUI (statistics tab) where it shows, for a single 
XDS_ASCII.HKL, how much batches of the run contribute towards CC1/2 of the 
dataset, i.e. what there delta-CC1/2 is. This e.g. allows to see where 
radiation damage sets in.

For use with merging of datasets, you create a XSCALE.INP with 

OUTPUT_FILE=WHATEVERNAME.HKL
INPUT_FILE=/some/where/first/XDS_ASCII.HKL
INPUT_FILE=/some/where/econd/XDS_ASCII.HKL
INPUT_FILE=/some/where/third/XDS_ASCII.HKL
... (as many INPUT_FILEs as you want)

Then you just run

xscale_par

and obtain XSCALE.LP and WHATEVERNAME.HKL (and a bunch of control files; if you 
are not interested in these, there are options that prevent their creation).

Finally, you run

xdscc12 WHATEVERNAME.HKL | tee XDSCC12.LP

and it will calculate the delta-CC1/2 statistics of each dataset, for each 
resolution shell.
It does this for the isomorphous signal (which is what you ask about), and for 
the anomalous signal (delta-CC1/2ano). 

There are options that modify the action and output of the program; you see 
them in the XDSwiki article or when you run with the -h option.

What should make its use easy is that XDSCC12 writes a file 
XSCALE.INP_rename_me that has the format of XSCALE.INP, but with the INPUT_FILE 
lines sorted by the quality of the datasets i.e. by their delta-CC1/2 . I'd 
read and edit XSCALE.INP_rename_me, remove the last one if it has a negative 
delta-CC1/2 , then rename to XSCALE.INP, and re-run xscale_par . If you have 
many datasets, this can be iterated; I hope you get the idea.

If you have questions, pls get back.

Hope this helps,
Kay





On Fri, 21 Jun 2024 15:15:34 +, Jordi Benach  wrote:

>Dear Everyone,
>Suppose you have three or more datasets and would like to choose the best 
>combination to merge and produce a higher quality dataset. I have been using 
>BLEND with mixed results. Additionally, BLEND doesn’t seem to accept datasets 
>with several runs (i.e., a dataset where several of its frames have been 
>removed, see next line).
>BLEND: “crystal_flag = 2:   crystal is rejected because data file is made up 
>of multiple runs”
>Apart from BLEND, what tools or methods do people use to achieve this? Using 
>CAD, SCALEIT, and/or POINTLESS should be helpful, but I haven’t found a 
>program that uses them to suggest the best dataset combination to merge.
>Thank you in advance for your help.
>Best regards,
>Jordi
>Jordi Benach, PhD
>LRL-CAT, Director & Head of Operations
>Sr. Director-Chemistry, DCRT
>Eli Lilly and Company
>Advanced Photon Source, Argonne National Laboratory, Building 438A
>9700 S. Cass Ave., Lemont, IL 60439 U.S.A.
>+1 630.252.0821 (office) | +1 630.209.4264 (mobile)
>benach_jo...@lilly.com | 
>www.lilly.com
>[cid:image001.png@01DAC3C3.F2C1B0F0]
>CONFIDENTIALITY NOTICE: This email message (including all attachments) is for 
>the sole use of the intended recipient(s) and may contain confidential 
>information. Any unauthorized review, use, disclosure, copying or distribution 
>is strictly prohibited. If you are not the intended recipient, please contact 
>the sender by reply email and destroy all copies of the original message.
>
>
>
>
>
>
>
>To unsubscribe from the CCP4BB list, click the following link:
>https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
>
>This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing 
>list hosted by www.jiscmail.ac.uk, terms & conditions are available at 
>https://www.jiscmail.ac.uk/policyandsecurity/



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/

Re: [ccp4bb] merging of datasets

[Eleanor Dodson]

I second martin’s suggestion..

it is not essential but it is sensible to give the lively best as the first
entry then the other sets are scaled and indexed to match it..
I use ccp4i2 for this . You will get quite a voluminous report which is
very helpful

Good luck Eleanor


On Fri, 21 Jun 2024 at 17:26, Martin Malý  wrote:

> Dear Jordi,
>
> For this purpose, I would use AIMLESS (you can specify a list of unmerged
> data sets) or xia2.multiplex. Output log files should indicate which data
> sets or images to discard - then the merging can be re-run without them.
>
> Best regards,
> Martin
>
>
> On 21/06/2024 16:15, Jordi Benach wrote:
>
> Dear Everyone,
>
> Suppose you have three or more datasets and would like to choose the best
> combination to merge and produce a higher quality dataset. I have been
> using BLEND with mixed results. Additionally, BLEND doesn’t seem to accept
> datasets with several runs (i.e., a dataset where several of its frames
> have been removed, see next line).
>
> BLEND: “crystal_flag = 2:   crystal is rejected because data file is made
> up of multiple runs”
>
> Apart from BLEND, what tools or methods do people use to achieve this?
> Using CAD, SCALEIT, and/or POINTLESS should be helpful, but I haven’t found
> a program that uses *them* to suggest the best dataset combination to
> merge.
>
> Thank you in advance for your help.
>
> Best regards,
>
> Jordi
>
> *Jordi Benach, PhD*
> LRL-CAT, Director & Head of Operations
>
> Sr. Director-Chemistry, DCRT
> Eli Lilly and Company
>
> Advanced Photon Source, Argonne National Laboratory, Building 438A
>
> 9700 S. Cass Ave., Lemont, IL 60439 U.S.A.
> 
>
> +1 630.252.0821 (office) | +1 630.209.4264 (mobile)
>
> benach_jo...@lilly.com | www.lilly.com
>
> CONFIDENTIALITY NOTICE: This email message (including all attachments) is
> for the sole use of the intended recipient(s) and may contain confidential
> information. Any unauthorized review, use, disclosure, copying or
> distribution is strictly prohibited. If you are not the intended recipient,
> please contact the sender by reply email and destroy all copies of the
> original message.
>
>
>
>
>
>
>
>
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
>
>
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/

Re: [ccp4bb] merging of datasets

[Martin Malý]

Dear Jordi,

For this purpose, I would use AIMLESS (you can specify a list of 
unmerged data sets) or xia2.multiplex. Output log files should indicate 
which data sets or images to discard - then the merging can be re-run 
without them.


Best regards,
Martin

On 21/06/2024 16:15, Jordi Benach wrote:


Dear Everyone,

Suppose you have three or more datasets and would like to choose the 
best combination to merge and produce a higher quality dataset. I have 
been using BLEND with mixed results. Additionally, BLEND doesn’t seem 
to accept datasets with several runs (i.e., a dataset where several of 
its frames have been removed, see next line).


BLEND: “crystal_flag = 2:   crystal is rejected because data file is 
made up of multiple runs”


Apart from BLEND, what tools or methods do people use to achieve this? 
Using CAD, SCALEIT, and/or POINTLESS should be helpful, but I haven’t 
found a program that uses /them/ to suggest the best dataset 
combination to merge.


Thank you in advance for your help.

Best regards,

Jordi

*Jordi Benach, PhD*
LRL-CAT, Director & Head of Operations

Sr. Director-Chemistry, DCRT
Eli Lilly and Company

Advanced Photon Source, Argonne National Laboratory, Building 438A

9700 S. Cass Ave., Lemont, IL 60439 U.S.A.

+1 630.252.0821 (office) | +1 630.209.4264 (mobile)

benach_jo...@lilly.com | www.lilly.com 



CONFIDENTIALITY NOTICE: This email message (including all attachments) 
is for the sole use of the intended recipient(s) and may contain 
confidential information. Any unauthorized review, use, disclosure, 
copying or distribution is strictly prohibited. If you are not the 
intended recipient, please contact the sender by reply email and 
destroy all copies of the original message.





To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1 







To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/

[ccp4bb] merging of datasets

[Jordi Benach]

Dear Everyone,
Suppose you have three or more datasets and would like to choose the best 
combination to merge and produce a higher quality dataset. I have been using 
BLEND with mixed results. Additionally, BLEND doesn’t seem to accept datasets 
with several runs (i.e., a dataset where several of its frames have been 
removed, see next line).
BLEND: “crystal_flag = 2:   crystal is rejected because data file is made up of 
multiple runs”
Apart from BLEND, what tools or methods do people use to achieve this? Using 
CAD, SCALEIT, and/or POINTLESS should be helpful, but I haven’t found a program 
that uses them to suggest the best dataset combination to merge.
Thank you in advance for your help.
Best regards,
Jordi
Jordi Benach, PhD
LRL-CAT, Director & Head of Operations
Sr. Director-Chemistry, DCRT
Eli Lilly and Company
Advanced Photon Source, Argonne National Laboratory, Building 438A
9700 S. Cass Ave., Lemont, IL 60439 U.S.A.
+1 630.252.0821 (office) | +1 630.209.4264 (mobile)
benach_jo...@lilly.com | 
www.lilly.com
[cid:image001.png@01DAC3C3.F2C1B0F0]
CONFIDENTIALITY NOTICE: This email message (including all attachments) is for 
the sole use of the intended recipient(s) and may contain confidential 
information. Any unauthorized review, use, disclosure, copying or distribution 
is strictly prohibited. If you are not the intended recipient, please contact 
the sender by reply email and destroy all copies of the original message.







To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/

Re: [ccp4bb] Extra density on Cysteine

[Mark J. van Raaij]

Dear Tina,

my guess would be minor oxidation of the cysteine. The green positive 
difference density suggests a bound atom in three different positions; this 
could however just be the first atom of something bigger like 
beta-mercaptoethanol, with the rest of that molecule too disordered to see. By 
eye-balling, the blue electron density does not look much bigger than expected 
for sulfur, so whatever it is, I would think it's very minor and probably not 
even worth modelling. 

I am also guessing this map is after refinement is almost complete and the 
overall difference map quite flat, so three sigmas of difference density (if 
that is what is contoured) may correspond to quite few electrons, i.e. low 
occupancy of the adduct.

Another explanation might be "anomalous behaviour" of the sulfur, similar to 
residual difference densities often observed near metal ions. A physicist will 
likely have a better idea about this than me.

Mark van Raaij
Dpto de Estructura de Macromoleculas, lab 20B
Centro Nacional de Biotecnologia - CSIC
calle Darwin 3
E-28049 Madrid, Spain
tel. +34 91 585 4616 (internal 432092)


> On 21 Jun 2024, at 11:02, hsyu11gmail  wrote:
> 
> Dear all,
> 
> We recently solved a structure to 2.1 A, and found an additional Fo-Fc 
> density on the side chain of a cysteine residue. The structure has been 
> reported previously, and no modification found at this site. The distances 
> between the cysteine and other residues also does not appear to be long 
> enough for any modifications. 
> 
> Does anyone has an idea about this map?
> 
> Thanks.
> 
> Kind regards,
> Tina
> 
> 
> 
> 
> hsyu11gmail
> 
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1
> 




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/

[ccp4bb] Research group leader opening at EMBL-EBI

[Sameer Velankar]

Dear All,

At EMBL-EBI, we are recruiting Group Leaders (Faculty members), with a closing 
date of July 14th and interviews on 02/03 September:
https://www.ebi.ac.uk/research/research-group-leaders-at-embl-ebi

EMBL-EBI, based near Cambridge, is a European Centre for Computational Biology. 
Our research faculty are often scientists who focus on developing and applying 
methods for biological data analysis and interpretation. Candidates driven to 
explore particular aspects of biology are also welcome.  EMBL-EBI has excellent 
computational capacities and hosts major resources, including PDBe, EMDB and 
AlphaFold Protein Structure Database. 

EMBL Group Leader positions are well suited for early-career applicants, from 
their first postdoctoral appointment or even the last year of their PhD. We 
select people based on their potential to become scientific leaders. 

For further information, please contact Julio Saez-Rodriguez, Head of Research 
at EMBL-EBI (https://www.ebi.ac.uk/people/person/julio-saez-rodriguez/).

Best wishes,
Sameer


To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB&A=1

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & conditions are available at 
https://www.jiscmail.ac.uk/policyandsecurity/