Re: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB -- N-glycans are now separate chains if more than one residue

[Jan Dohnalek]

We also recently encountered this way of coordinates treatment - did not
find it useful but just left it as it was (life has other more exciting
adventures waiting ...).

I strongly support the solution suggested by Luca,

A nice day to all,
Jan Dohnalek


On Fri, Dec 4, 2020 at 8:48 AM Luca Jovine  wrote:

> CC: pdb-l
>
> Dear Zhijie and Robbie,
>
> I agree with both of you that the new carbohydrate chain assignment
> convention that has been recently adopted by PDB introduces confusion, not
> just for PDB-REDO but also - and especially - for end users.
>
> Could we kindly ask PDB to improve consistency by either assigning a
> separate chain to all covalently attached carbohydrates (regardless of
> whether one or more residues have been traced), or reverting to the old
> system (where N-/O-glycans inherited the same chain ID of the protein to
> which they are attached)? The current hybrid solution hardly seems
> optimal...
>
> Best regards,
>
> Luca
>
> > On 3 Dec 2020, at 20:17, Robbie Joosten 
> wrote:
> >
> > Dear Zhijie,
> >
> > In generally I like the treatment of carbohydrates now as branched
> polymers. I didn't realise there was an exception. It makes sense for
> unlinked carbohydrate ligands, but not for N- or O-glycosylation sites as
> these might change during model building or, in my case, carbohydrate
> rebuilding in PDB-REDO powered by Coot. Thanks for pointing this out.
> >
> > Cheers,
> > Robbie
> >
> >> -Original Message-
> >> From: CCP4 bulletin board  On Behalf Of Zhijie
> Li
> >> Sent: Thursday, December 3, 2020 19:52
> >> To: CCP4BB@JISCMAIL.AC.UK
> >> Subject: Re: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the
> >> PDB -- N-glycans are now separate chains if more than one residue
> >>
> >> Hi all,
> >>
> >> I was confused when I saw mysterious new glycan chains emerging during
> >> PDB deposition and spent quite some time trying to find out what was
> >> wrong with my coordinates.  Then it occurred to me that a lot of recent
> >> structures also had tens of N-glycan chains.  Finally I realized that
> this
> >> phenomenon is a consequence of this PDB policy announced here in July.
> >>
> >>
> >> For future depositors who might also get puzzled, let's put it in a
> short
> >> sentence:  O- and N-glycans are now separate chains if it they contain
> more
> >> than one residue; single residues remain with the protein chain.
> >>
> >>
> >>
> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.wwpdb.org%2Fdocumentation%2Fcarbohydrate-remediationdata=04%7C01%7Cluca.jovine%40KI.SE%7C1d790a0717ce4217c7a308d897c01b47%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C1%7C637426199684263065%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000sdata=mBrkCJECFpZyCih4kOCcCvLT1GzQHxD5GD7bZDI9s1s%3Dreserved=0
> >>
> >> "Oligosaccharide molecules are classified as a new entity type,
> branched,
> >> assigned a unique chain ID (_atom_site.auth_asym_id) and a new mmCIF
> >> category introduced to define the type of branching
> >> (_pdbx_entity_branch.type) . "
> >>
> >>
> >>
> >>
> >>
> >> I found the differential treatment of single-residue glycans and
> multi-residue
> >> glycans not only bit lack of aesthetics but also misleading.  When a
> structure
> >> contains both NAG-NAG... and single NAG on N-glycosylation sites, it
> might
> >> be because of lack of density for building more residues, or because
> that
> >> some of the glycosylation sites are now indeed single NAGs (endoH etc.)
> >> while some others are not cleaved due to accessibility issues.
> Leaving NAGs
> >> on the protein chain while assigning NAG-NAG... to a new chain, feels
> like
> >> suggesting something about their true oligomeric state.
> >>
> >>
> >> For example, for cryoEM structures, when one only builds a single NAG
> at a
> >> site does not necessarily mean that the protein was treated by endoH. In
> >> fact all sites are extended to at least tri-Man in most cases. Then why
> >> keeping some sites associated with the protein chain while others kicked
> >> out?
> >>
> >> Zhijie
> >>
> >>
> >>
> >> 
> >>
> >> From: CCP4 bulletin board  on behalf of John
> >> Berrisford 
> >> Sent: Thursday, July 9, 2020 4:39 AM
> >> To: CCP4BB@JISCMAIL.AC.UK 
> >> Subject: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB
> >>
> >>
> >> Dear CCP4BB
> >>
> >> PDB data will shortly incorporate a new data representation for
> >> carbohydrates in PDB entries and reference data that improves the
> >> Findability and Interoperability of these molecules in macromolecular
> >> structures. In order to remediate and improve the representation of
> >> carbohydrates across the archive, the wwPDB has:
> >>
> >> *standardized Chemical Component Dictionary nomenclature
> >> following IUPAC-IUBMB recommendations
> >> *provided uniform representation for oligosaccharides
> >> *adopted Glycoscience-community commonly used linear descriptors
> >> 

Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

[Joana Pereira]

Hi everybody,


As one of the persons playing with the CASP14 data before all news came 
out, I can answer some of the questions raised in this thread.



- "Does anyone know how AlphaFold performs on sequences with little 
conservation?"


One of the things we looked at was how the accuracy of the models was 
dependent on the Neff (number of effective sequences, relates to how 
deep alignments are for that sequence and, thus, to the number of 
homologs and the conservation of the sequence). What we could see is 
that, basically, in CASP14 it does not anymore and that (near-)singleton 
sequences could be modeled with a pretty good accuracy.



- "It would be interesting to know how it performs with structures of 
new or uncertain fold."


It does pretty well! Similarly to the Neff relationship, we also see a 
basically flat line at a GDT of 70-80 at any level of target difficulty. 
Of course the accuracy is slightly higher for easy targets (those for 
which there are templates in the PDB), but to have a GDT of around 70 in 
Free-Modelling, hard targets, is quite impressive.



- "I don't think they have all the side chain placement so perfect as to 
be able to predict the fold _and_ how a compound or another protein binds"


Yap, sidechains remain the poorest modeled parts. Still, those modeled 
by AlphaFold were the closest to the "reality" of the target...



- "I'm curious how well AlphaFold would do on an*Intrinsically 
Disordered Protein (IDP)*"


Oh yes, that is a super good point and I have been thinking about it 
too. Maybe one should start throwing some IDPs into CASP too :) There's 
the CAID experiment but, on its current state, AlphaFold would not be 
possible to test.



Best wishes

Joana


---

Dr. Joana Pereira
Postdoctoral Researcher
Department of Protein Evolution

Max Planck Institute for Developmental Biology
Max-Planck-Ring 5
72076 Tübingen
GERMANY



On 03.12.20 23:46, Reza Khayat wrote:


​Does anyone know how AlphaFold performs on sequences with little 
conservation? Virus and phage proteins are like this. Their structures 
are homologous, but sequence identity can be less than 10%.



Reza


Reza Khayat, PhD
Associate Professor
City College of New York
Department of Chemistry and Biochemistry
New York, NY 10031

*From:* CCP4 bulletin board  on behalf of 
Anastassis Perrakis 

*Sent:* Thursday, December 3, 2020 5:31 PM
*To:* CCP4BB@JISCMAIL.AC.UK
*Subject:* [EXTERNAL] Re: [ccp4bb] AlphaFold: more thinking and less 
pipetting (?)
AlphaFold - or similar ideas that will surface up sooner or later - 
will beyond doubt have major impact. The accuracy it demonstrated 
compared to others is excellent.


“Our” target (T1068) that was not solvable by MR with the homologous 
search structure or a homology model (it was phased with Archimboldo, 
rather easily), is easily solvable with the AlphaFold model as a 
search model. In PHASER I get Rotation Z-score 17.9, translation 
Z-score 26.0, using defaults.



imho what remains to be seen is:

a. how and when will a prediction server be available?
b. even if training needs computing that will surely unaccessible to 
most, will there be code that can be installed in a “reasonable” 
number of GPUs and how fast will it be?
c. how do model quality metrics (that do not compared with the known 
answer) correlate with the expected RMSD? AlphaFold, no matter how 
impressive, still gets things wrong.
c. will the AI efforts now gear to ligand (fragment?) prediction with 
similarly impressive performance?


Exciting times.

A.




On 3 Dec 2020, at 21:55, Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk 
> wrote:


Hello. A quick look suggests that a lot of the test structures were 
solved by phaser or molrep, suggesting it is a very welcome 
improvement on homology modelling. It would be interesting to know 
how it performs with structures of new or uncertain fold, if there 
are any left these days. Without resorting to jokes about artificial 
intelligence, I couldn't make that out from the CASP14 website or the 
many excellent articles that have appeared. Best wishes, Jon Cooper.



Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < isabel.gar...@ibs.fr 
> wrote:



Dear all,

Just commenting that after the stunning performance of AlphaFold
that uses AI from Google maybe some of us we could dedicate
ourselves to the noble art of gardening, baking, doing Chinese
Calligraphy, enjoying the clouds pass or everything together
(just in case I have already prepared my subscription to Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Re: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB -- N-glycans are now separate chains if more than one residue

[Luca Jovine]

CC: pdb-l

Dear Zhijie and Robbie,

I agree with both of you that the new carbohydrate chain assignment convention 
that has been recently adopted by PDB introduces confusion, not just for 
PDB-REDO but also - and especially - for end users.

Could we kindly ask PDB to improve consistency by either assigning a separate 
chain to all covalently attached carbohydrates (regardless of whether one or 
more residues have been traced), or reverting to the old system (where 
N-/O-glycans inherited the same chain ID of the protein to which they are 
attached)? The current hybrid solution hardly seems optimal...

Best regards,

Luca

> On 3 Dec 2020, at 20:17, Robbie Joosten  wrote:
>
> Dear Zhijie,
>
> In generally I like the treatment of carbohydrates now as branched polymers. 
> I didn't realise there was an exception. It makes sense for unlinked 
> carbohydrate ligands, but not for N- or O-glycosylation sites as these might 
> change during model building or, in my case, carbohydrate rebuilding in 
> PDB-REDO powered by Coot. Thanks for pointing this out.
>
> Cheers,
> Robbie
>
>> -Original Message-
>> From: CCP4 bulletin board  On Behalf Of Zhijie Li
>> Sent: Thursday, December 3, 2020 19:52
>> To: CCP4BB@JISCMAIL.AC.UK
>> Subject: Re: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the
>> PDB -- N-glycans are now separate chains if more than one residue
>>
>> Hi all,
>>
>> I was confused when I saw mysterious new glycan chains emerging during
>> PDB deposition and spent quite some time trying to find out what was
>> wrong with my coordinates.  Then it occurred to me that a lot of recent
>> structures also had tens of N-glycan chains.  Finally I realized that this
>> phenomenon is a consequence of this PDB policy announced here in July.
>>
>>
>> For future depositors who might also get puzzled, let's put it in a short
>> sentence:  O- and N-glycans are now separate chains if it they contain more
>> than one residue; single residues remain with the protein chain.
>>
>>
>> https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fwww.wwpdb.org%2Fdocumentation%2Fcarbohydrate-remediationdata=04%7C01%7Cluca.jovine%40KI.SE%7C1d790a0717ce4217c7a308d897c01b47%7Cbff7eef1cf4b4f32be3da1dda043c05d%7C0%7C1%7C637426199684263065%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C1000sdata=mBrkCJECFpZyCih4kOCcCvLT1GzQHxD5GD7bZDI9s1s%3Dreserved=0
>>
>> "Oligosaccharide molecules are classified as a new entity type, branched,
>> assigned a unique chain ID (_atom_site.auth_asym_id) and a new mmCIF
>> category introduced to define the type of branching
>> (_pdbx_entity_branch.type) . "
>>
>>
>>
>>
>>
>> I found the differential treatment of single-residue glycans and 
>> multi-residue
>> glycans not only bit lack of aesthetics but also misleading.  When a 
>> structure
>> contains both NAG-NAG... and single NAG on N-glycosylation sites, it might
>> be because of lack of density for building more residues, or because that
>> some of the glycosylation sites are now indeed single NAGs (endoH etc.)
>> while some others are not cleaved due to accessibility issues.Leaving 
>> NAGs
>> on the protein chain while assigning NAG-NAG... to a new chain, feels like
>> suggesting something about their true oligomeric state.
>>
>>
>> For example, for cryoEM structures, when one only builds a single NAG at a
>> site does not necessarily mean that the protein was treated by endoH. In
>> fact all sites are extended to at least tri-Man in most cases. Then why
>> keeping some sites associated with the protein chain while others kicked
>> out?
>>
>> Zhijie
>>
>>
>>
>> 
>>
>> From: CCP4 bulletin board  on behalf of John
>> Berrisford 
>> Sent: Thursday, July 9, 2020 4:39 AM
>> To: CCP4BB@JISCMAIL.AC.UK 
>> Subject: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB
>>
>>
>> Dear CCP4BB
>>
>> PDB data will shortly incorporate a new data representation for
>> carbohydrates in PDB entries and reference data that improves the
>> Findability and Interoperability of these molecules in macromolecular
>> structures. In order to remediate and improve the representation of
>> carbohydrates across the archive, the wwPDB has:
>>
>> *standardized Chemical Component Dictionary nomenclature
>> following IUPAC-IUBMB recommendations
>> *provided uniform representation for oligosaccharides
>> *adopted Glycoscience-community commonly used linear descriptors
>> using community tools
>> *annotated glycosylation sites in PDB structures
>>
>> Starting July 29, 2020, users will be able to access the improved data via 
>> FTP
>> or wwPDB partner websites. Detailed information about this project is
>> available at the wwPDB website
>> 

Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

[Joel Sussman]

I'm curious how well AlphaFold would do on an Intrinsically Disordered Protein 
(IDP),
would it recognize that it is an "IDP" or predict that it has a structure (or 
structures)?
It would be interesting to test such a sequence and see what comes out.
Possibly AlphaFold might be the best IDP predictor too.

Joel


On 4 Dec 2020, at 6:29, Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk>
 wrote:

Hello James, that's really strange - I've used refmac et al., to do poor man's 
energy minimizations of models and they've generally come out fine, unless the 
restraints, etc, are wildly off-target. I wasn't playing with X-ray weights 
though, since there never was a dataset, of course.

Cheers, Jon.C.

Sent from ProtonMail mobile



 Original Message 
On 4 Dec 2020, 01:34, James Holton < jmhol...@lbl.gov> 
wrote:

It is a major leap forward for structure prediction for sure.  A hearty 
congratulations to all those teams over all those years.

The part I don't understand is the accuracy.  If we understand what holds 
molecules together so well, then why is it that when I refine an X-ray 
structure and turn the X-ray weight term down to zero ... the molecule blows up 
in my face?

-James Holton
MAD Scientist


On 12/3/2020 3:17 AM, Isabel Garcia-Saez wrote:
Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times.

Cheers,

Isabel


Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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




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



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




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Jon Cooper]

Hello James, that's really strange - I've used refmac et al., to do poor man's 
energy minimizations of models and they've generally come out fine, unless the 
restraints, etc, are wildly off-target. I wasn't playing with X-ray weights 
though, since there never was a dataset, of course.

Cheers, Jon.C.

Sent from ProtonMail mobile

 Original Message 
On 4 Dec 2020, 01:34, James Holton wrote:

> It is a major leap forward for structure prediction for sure. A hearty 
> congratulations to all those teams over all those years.
>
> The part I don't understand is the accuracy. If we understand what holds 
> molecules together so well, then why is it that when I refine an X-ray 
> structure and turn the X-ray weight term down to zero ... the molecule blows 
> up in my face?
>
> -James Holton
> MAD Scientist
>
> On 12/3/2020 3:17 AM, Isabel Garcia-Saez wrote:
>
>> Dear all,
>>
>> Just commenting that after the stunning performance of AlphaFold that uses 
>> AI from Google maybe some of us we could dedicate ourselves to the noble art 
>> of gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
>> everything together (just in case I have already prepared my subscription to 
>> Netflix).
>>
>> https://www.nature.com/articles/d41586-020-03348-4
>>
>> Well, I suppose that we still have the structures of complexes (at the 
>> moment). I am wondering how the labs will have access to this technology in 
>> the future (would it be for free coming from the company DeepMind - 
>> Google?). It seems that they have already published some code. Well, 
>> exciting times.
>>
>> Cheers,
>>
>> Isabel
>>
>> Isabel Garcia-SaezPhD
>> Institut de Biologie Structurale
>> Viral Infection and Cancer Group (VIC)-Cell Division Team
>> 71, Avenue des Martyrs
>> CS 10090
>> 38044 Grenoble Cedex 9
>> France
>> Tel.: 00 33 (0) 457 42 86 15
>> [e-mail: isabel.gar...@ibs.fr](mailto:isabel.gar...@ibs.fr)
>> FAX: 00 33 (0) 476 50 18 90
>> http://www.ibs.fr/
>>
>> ---
>>
>> To unsubscribe from the CCP4BB list, click the following link:
>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
>
> ---
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[James Holton]

It is a major leap forward for structure prediction for sure.  A hearty 
congratulations to all those teams over all those years.


The part I don't understand is the accuracy.  If we understand what 
holds molecules together so well, then why is it that when I refine an 
X-ray structure and turn the X-ray weight term down to zero ... the 
molecule blows up in my face?


-James Holton
MAD Scientist


On 12/3/2020 3:17 AM, Isabel Garcia-Saez wrote:

Dear all,

Just commenting that after the stunning performance of AlphaFold that 
uses AI from Google maybe some of us we could dedicate ourselves to 
the noble art of gardening, baking, doing Chinese Calligraphy, 
enjoying the clouds pass or everything together (just in case I have 
already prepared my subscription to Netflix).


https://www.nature.com/articles/d41586-020-03348-4 



Well, I suppose that we still have the structures of complexes (at the 
moment). I am wondering how the labs will have access to this 
technology in the future (would it be for free coming from the company 
DeepMind - Google?). It seems that they have already published some 
code. Well, exciting times.


Cheers,

Isabel


Isabel Garcia-SaezPhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr 
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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








To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Paul Adams]

I agree completely Tom. Having been recently involved in some efforts to 
identify interesting compounds against SARS-CoV-2, I can say that the current 
AI/ML methods for docking/predicting small molecule binding have very very low 
success rates (I’m being generous here), even when you are working with the 
experimental protein structure! Maybe this is the next frontier for the 
prediction methods (after they’ve solved the protein/protein complex problem of 
course), but it seems there is a long way to go.

Given that many structures are solved to look at their interaction with other 
proteins or small molecules I think that experimental structural biology is 
here to stay for a while - past Tom’s retirement even! However, will these 
fairly accurate protein predictions make experimental phasing a thing of the 
past?


> On Dec 3, 2020, at 4:16 PM, Peat, Tom (Manufacturing, Parkville) 
>  wrote:
> 
> Although they can now get the fold correct, I don't think they have all the 
> side chain placement so perfect as to be able to predict the fold and how a 
> compound or another protein binds, so we can still do complexes. I don't know 
> what others end up spending their time doing, but much of my work has been 
> trying to fit ligands into density, which may take another few years of 
> algorithm development, which is fine for me as I can retire! 
> cheers, tom 
> 
> Tom Peat, PhD
> Proteins Group
> Biomedical Program, CSIRO
> 343 Royal Parade
> Parkville, VIC, 3052
> +613 9662 7304
> +614 57 539 419
> tom.p...@csiro.au 
> 
> From: CCP4 bulletin board  > on behalf of Jon Cooper 
> <488a26d62010-dmarc-requ...@jiscmail.ac.uk 
> >
> Sent: Friday, December 4, 2020 9:55 AM
> To: CCP4BB@JISCMAIL.AC.UK  
> mailto:CCP4BB@JISCMAIL.AC.UK>>
> Subject: Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)
>  
> Thanks all, very interesting, so our methods are just needed to identify the 
> crystallization impurities, when the trays have been thrown away ;-
> 
> Cheers, Jon.C.
> 
> Sent from ProtonMail mobile
> 
> 
> 
>  Original Message 
> On 3 Dec 2020, 22:31, Anastassis Perrakis < a.perra...@nki.nl 
> > wrote:
> 
> AlphaFold - or similar ideas that will surface up sooner or later - will 
> beyond doubt have major impact. The accuracy it demonstrated compared to 
> others is excellent.
> 
> “Our” target (T1068) that was not solvable by MR with the homologous search 
> structure or a homology model (it was phased with Archimboldo, rather 
> easily), is easily solvable with the AlphaFold model as a search model. In 
> PHASER I get Rotation Z-score 17.9, translation Z-score 26.0, using defaults.
> 
> 
> imho what remains to be seen is:
> 
> a. how and when will a prediction server be available? 
> b. even if training needs computing that will surely unaccessible to most, 
> will there be code that can be installed in a “reasonable” number of GPUs and 
> how fast will it be?
> c. how do model quality metrics (that do not compared with the known answer) 
> correlate with the expected RMSD? AlphaFold, no matter how impressive, still 
> gets things wrong.
> c. will the AI efforts now gear to ligand (fragment?) prediction with 
> similarly impressive performance?
> 
> Exciting times.
> 
> A.
> 
> 
> 
> 
>> On 3 Dec 2020, at 21:55, Jon Cooper 
>> <488a26d62010-dmarc-requ...@jiscmail.ac.uk 
>> > wrote:
>> 
>> Hello. A quick look suggests that a lot of the test structures were solved 
>> by phaser or molrep, suggesting it is a very welcome improvement on homology 
>> modelling. It would be interesting to know how it performs with structures 
>> of new or uncertain fold, if there are any left these days. Without 
>> resorting to jokes about artificial intelligence, I couldn't make that out 
>> from the CASP14 website or the many excellent articles that have appeared. 
>> Best wishes, Jon Cooper.
>> 
>> 
>> Sent from ProtonMail mobile
>> 
>> 
>> 
>>  Original Message 
>> On 3 Dec 2020, 11:17, Isabel Garcia-Saez < isabel.gar...@ibs.fr 
>> > wrote:
>> 
>> Dear all,
>> 
>> Just commenting that after the stunning performance of AlphaFold that uses 
>> AI from Google maybe some of us we could dedicate ourselves to the noble art 
>> of gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
>> everything together (just in case I have already prepared my subscription to 
>> Netflix).
>> 
>> https://www.nature.com/articles/d41586-020-03348-4 
>> 
>> 
>> Well, I suppose that we still have the structures of complexes (at the 
>> moment). I am wondering how the labs will have access to this technology in 
>> the future (would it be for free coming from the company DeepMind - 
>> 

Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

[Peat, Tom (Manufacturing, Parkville)]

Although they can now get the fold correct, I don't think they have all the 
side chain placement so perfect as to be able to predict the fold and how a 
compound or another protein binds, so we can still do complexes. I don't know 
what others end up spending their time doing, but much of my work has been 
trying to fit ligands into density, which may take another few years of 
algorithm development, which is fine for me as I can retire!
cheers, tom

Tom Peat, PhD
Proteins Group
Biomedical Program, CSIRO
343 Royal Parade
Parkville, VIC, 3052
+613 9662 7304
+614 57 539 419
tom.p...@csiro.au


From: CCP4 bulletin board  on behalf of Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk>
Sent: Friday, December 4, 2020 9:55 AM
To: CCP4BB@JISCMAIL.AC.UK 
Subject: Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

Thanks all, very interesting, so our methods are just needed to identify the 
crystallization impurities, when the trays have been thrown away ;-

Cheers, Jon.C.

Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 22:31, Anastassis Perrakis < a.perra...@nki.nl> wrote:

AlphaFold - or similar ideas that will surface up sooner or later - will beyond 
doubt have major impact. The accuracy it demonstrated compared to others is 
excellent.

“Our” target (T1068) that was not solvable by MR with the homologous search 
structure or a homology model (it was phased with Archimboldo, rather easily), 
is easily solvable with the AlphaFold model as a search model. In PHASER I get 
Rotation Z-score 17.9, translation Z-score 26.0, using defaults.


imho what remains to be seen is:

a. how and when will a prediction server be available?
b. even if training needs computing that will surely unaccessible to most, will 
there be code that can be installed in a “reasonable” number of GPUs and how 
fast will it be?
c. how do model quality metrics (that do not compared with the known answer) 
correlate with the expected RMSD? AlphaFold, no matter how impressive, still 
gets things wrong.
c. will the AI efforts now gear to ligand (fragment?) prediction with similarly 
impressive performance?

Exciting times.

A.




On 3 Dec 2020, at 21:55, Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk>
 wrote:

Hello. A quick look suggests that a lot of the test structures were solved by 
phaser or molrep, suggesting it is a very welcome improvement on homology 
modelling. It would be interesting to know how it performs with structures of 
new or uncertain fold, if there are any left these days. Without resorting to 
jokes about artificial intelligence, I couldn't make that out from the CASP14 
website or the many excellent articles that have appeared. Best wishes, Jon 
Cooper.


Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < 
isabel.gar...@ibs.fr> wrote:

Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times.

Cheers,

Isabel


Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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



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




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



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



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

This message was issued to members of www.jiscmail.ac.uk/CCP4BB, a mailing list 
hosted by www.jiscmail.ac.uk, terms & 

Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

[Boaz Shaanan]

Just curious, how does the result of the Phaser run  with the Alphafold model compare with a Phaser run using the Arcimboldo phased model as a probe?
Boaz

Boaz Shaanan, Ph.D.
Department of Life Sciences
Ben Gurion University of the Negev
Beer Sheva
Israel



On Dec 4, 2020 00:32, Anastassis Perrakis  wrote:


AlphaFold - or similar ideas that will surface up sooner or later - will beyond doubt have major impact. The accuracy it demonstrated compared to others is excellent.


“Our” target (T1068) that was not solvable by MR with the homologous search structure or a homology model (it was phased with Archimboldo, rather easily), is easily solvable with the AlphaFold model as a search model. In PHASER I get Rotation
 Z-score 17.9, translation Z-score 26.0, using defaults.




imho what remains to be seen is:


a. how and when will a prediction server be available? 
b. even if training needs computing that will surely unaccessible to most, will there be code that can be installed in a “reasonable” number of GPUs and how fast will it be?
c. how do model quality metrics (that do not compared with the known answer) correlate with the expected RMSD? AlphaFold, no matter how impressive, still gets things wrong.
c. will the AI efforts now gear to ligand (fragment?) prediction with similarly impressive performance?


Exciting times.


A.










On 3 Dec 2020, at 21:55, Jon Cooper <488a26d62010-dmarc-requ...@jiscmail.ac.uk> wrote:

Hello. A quick look suggests that a lot of the test structures were solved by phaser or molrep, suggesting it is a very welcome improvement on homology modelling. It would be interesting to know how it performs with structures of new or uncertain
 fold, if there are any left these days. Without resorting to jokes about artificial intelligence, I couldn't make that out from the CASP14 website or the many excellent articles that have appeared. Best wishes, Jon Cooper.


Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < 
isabel.gar...@ibs.fr> wrote:


Dear all,


Just commenting that after the stunning performance of AlphaFold that uses AI from Google maybe some of us we could dedicate ourselves to the noble art of gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or everything together
 (just in case I have already prepared my subscription to Netflix).


https://www.nature.com/articles/d41586-020-03348-4


Well, I suppose that we still have the structures of complexes (at the moment). I am wondering how the labs will have access to this technology in the future (would it be for free coming from the company DeepMind - Google?). It seems that they
 have already published some code. Well, exciting times. 


Cheers,


Isabel







Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/





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



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










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






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

Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

[Jon Cooper]

Thanks all, very interesting, so our methods are just needed to identify the 
crystallization impurities, when the trays have been thrown away ;-

Cheers, Jon.C.

Sent from ProtonMail mobile

 Original Message 
On 3 Dec 2020, 22:31, Anastassis Perrakis wrote:

> AlphaFold - or similar ideas that will surface up sooner or later - will 
> beyond doubt have major impact. The accuracy it demonstrated compared to 
> others is excellent.
>
> “Our” target (T1068) that was not solvable by MR with the homologous search 
> structure or a homology model (it was phased with Archimboldo, rather 
> easily), is easily solvable with the AlphaFold model as a search model. In 
> PHASER I get Rotation Z-score 17.9, translation Z-score 26.0, using defaults.
>
> imho what remains to be seen is:
>
> a. how and when will a prediction server be available?
> b. even if training needs computing that will surely unaccessible to most, 
> will there be code that can be installed in a “reasonable” number of GPUs and 
> how fast will it be?
> c. how do model quality metrics (that do not compared with the known answer) 
> correlate with the expected RMSD? AlphaFold, no matter how impressive, still 
> gets things wrong.
> c. will the AI efforts now gear to ligand (fragment?) prediction with 
> similarly impressive performance?
>
> Exciting times.
>
> A.
>
>> On 3 Dec 2020, at 21:55, Jon Cooper 
>> <488a26d62010-dmarc-requ...@jiscmail.ac.uk> wrote:
>>
>> Hello. A quick look suggests that a lot of the test structures were solved 
>> by phaser or molrep, suggesting it is a very welcome improvement on homology 
>> modelling. It would be interesting to know how it performs with structures 
>> of new or uncertain fold, if there are any left these days. Without 
>> resorting to jokes about artificial intelligence, I couldn't make that out 
>> from the CASP14 website or the many excellent articles that have appeared. 
>> Best wishes, Jon Cooper.
>>
>> Sent from ProtonMail mobile
>>
>>  Original Message 
>> On 3 Dec 2020, 11:17, Isabel Garcia-Saez < isabel.gar...@ibs.fr> wrote:
>>
>>> Dear all,
>>>
>>> Just commenting that after the stunning performance of AlphaFold that uses 
>>> AI from Google maybe some of us we could dedicate ourselves to the noble 
>>> art of gardening, baking, doing Chinese Calligraphy, enjoying the clouds 
>>> pass or everything together (just in case I have already prepared my 
>>> subscription to Netflix).
>>>
>>> https://www.nature.com/articles/d41586-020-03348-4
>>>
>>> Well, I suppose that we still have the structures of complexes (at the 
>>> moment). I am wondering how the labs will have access to this technology in 
>>> the future (would it be for free coming from the company DeepMind - 
>>> Google?). It seems that they have already published some code. Well, 
>>> exciting times.
>>>
>>> Cheers,
>>>
>>> Isabel
>>>
>>> Isabel Garcia-Saez PhD
>>> Institut de Biologie Structurale
>>> Viral Infection and Cancer Group (VIC)-Cell Division Team
>>> 71, Avenue des Martyrs
>>> CS 10090
>>> 38044 Grenoble Cedex 9
>>> France
>>> Tel.: 00 33 (0) 457 42 86 15
>>> [e-mail: isabel.gar...@ibs.fr](mailto:isabel.gar...@ibs.fr)
>>> FAX: 00 33 (0) 476 50 18 90
>>> http://www.ibs.fr/
>>>
>>> ---
>>>
>>> To unsubscribe from the CCP4BB list, click the following link:
>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
>>>
>>> ---
>>>
>>> To unsubscribe from the CCP4BB list, click the following link:
>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
>
> ---
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Reza Khayat]

?Does anyone know how AlphaFold performs on sequences with little conservation? 
Virus and phage proteins are like this. Their structures are homologous, but 
sequence identity can be less than 10%.


Reza


Reza Khayat, PhD
Associate Professor
City College of New York
Department of Chemistry and Biochemistry
New York, NY 10031

From: CCP4 bulletin board  on behalf of Anastassis 
Perrakis 
Sent: Thursday, December 3, 2020 5:31 PM
To: CCP4BB@JISCMAIL.AC.UK
Subject: [EXTERNAL] Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

AlphaFold - or similar ideas that will surface up sooner or later - will beyond 
doubt have major impact. The accuracy it demonstrated compared to others is 
excellent.

"Our" target (T1068) that was not solvable by MR with the homologous search 
structure or a homology model (it was phased with Archimboldo, rather easily), 
is easily solvable with the AlphaFold model as a search model. In PHASER I get 
Rotation Z-score 17.9, translation Z-score 26.0, using defaults.


imho what remains to be seen is:

a. how and when will a prediction server be available?
b. even if training needs computing that will surely unaccessible to most, will 
there be code that can be installed in a "reasonable" number of GPUs and how 
fast will it be?
c. how do model quality metrics (that do not compared with the known answer) 
correlate with the expected RMSD? AlphaFold, no matter how impressive, still 
gets things wrong.
c. will the AI efforts now gear to ligand (fragment?) prediction with similarly 
impressive performance?

Exciting times.

A.




On 3 Dec 2020, at 21:55, Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk>
 wrote:

Hello. A quick look suggests that a lot of the test structures were solved by 
phaser or molrep, suggesting it is a very welcome improvement on homology 
modelling. It would be interesting to know how it performs with structures of 
new or uncertain fold, if there are any left these days. Without resorting to 
jokes about artificial intelligence, I couldn't make that out from the CASP14 
website or the many excellent articles that have appeared. Best wishes, Jon 
Cooper.


Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < 
isabel.gar...@ibs.fr> wrote:

Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times.

Cheers,

Isabel


Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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



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




To unsubscribe from the CCP4BB list, click the following 

Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

[Anastassis Perrakis]

AlphaFold - or similar ideas that will surface up sooner or later - will beyond 
doubt have major impact. The accuracy it demonstrated compared to others is 
excellent.

“Our” target (T1068) that was not solvable by MR with the homologous search 
structure or a homology model (it was phased with Archimboldo, rather easily), 
is easily solvable with the AlphaFold model as a search model. In PHASER I get 
Rotation Z-score 17.9, translation Z-score 26.0, using defaults.


imho what remains to be seen is:

a. how and when will a prediction server be available?
b. even if training needs computing that will surely unaccessible to most, will 
there be code that can be installed in a “reasonable” number of GPUs and how 
fast will it be?
c. how do model quality metrics (that do not compared with the known answer) 
correlate with the expected RMSD? AlphaFold, no matter how impressive, still 
gets things wrong.
c. will the AI efforts now gear to ligand (fragment?) prediction with similarly 
impressive performance?

Exciting times.

A.




On 3 Dec 2020, at 21:55, Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk>
 wrote:

Hello. A quick look suggests that a lot of the test structures were solved by 
phaser or molrep, suggesting it is a very welcome improvement on homology 
modelling. It would be interesting to know how it performs with structures of 
new or uncertain fold, if there are any left these days. Without resorting to 
jokes about artificial intelligence, I couldn't make that out from the CASP14 
website or the many excellent articles that have appeared. Best wishes, Jon 
Cooper.


Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < 
isabel.gar...@ibs.fr> wrote:

Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times.

Cheers,

Isabel


Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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



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




To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Peat, Tom (Manufacturing, Parkville)]

Hello Jon,

We had a novel structure and it did very well. I haven't tried the MR to see if 
the model would work, but it was so close that I can't imagine it not working. 
Our protein was ~185 residues and the closest PDB structures were about 4 
Angstrom rmsd over about 70 residues over just one part of the structure (the 
beta-sheet running through the middle), so pretty different to all the known 
structures.
So I agree with others, the predictions were much better this year.
cheers, tom

Tom Peat, PhD
Proteins Group
Biomedical Program, CSIRO
343 Royal Parade
Parkville, VIC, 3052
+613 9662 7304
+614 57 539 419
tom.p...@csiro.au


From: CCP4 bulletin board  on behalf of Jon Cooper 
<488a26d62010-dmarc-requ...@jiscmail.ac.uk>
Sent: Friday, December 4, 2020 7:55 AM
To: CCP4BB@JISCMAIL.AC.UK 
Subject: Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

Hello. A quick look suggests that a lot of the test structures were solved by 
phaser or molrep, suggesting it is a very welcome improvement on homology 
modelling. It would be interesting to know how it performs with structures of 
new or uncertain fold, if there are any left these days. Without resorting to 
jokes about artificial intelligence, I couldn't make that out from the CASP14 
website or the many excellent articles that have appeared. Best wishes, Jon 
Cooper.


Sent from ProtonMail mobile



 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < isabel.gar...@ibs.fr> wrote:

Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times.

Cheers,

Isabel


Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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



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



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Wim Burmeister]

Hello, 
we had a 124 aa target in Casp14, without any detectable homology to a known 
structure. Within the experimental errors, the AlphaFold2 model is identical to 
the NMR model we got. That was very convincing. 
Best wishes 
Wim 


De: "Jon Cooper" <488a26d62010-dmarc-requ...@jiscmail.ac.uk> 
À: "CCP4BB"  
Envoyé: Jeudi 3 Décembre 2020 21:55:38 
Objet: Re: [ccp4bb] AlphaFold: more thinking and less pipetting (?) 

Hello. A quick look suggests that a lot of the test structures were solved by 
phaser or molrep, suggesting it is a very welcome improvement on homology 
modelling. It would be interesting to know how it performs with structures of 
new or uncertain fold, if there are any left these days. Without resorting to 
jokes about artificial intelligence, I couldn't make that out from the CASP14 
website or the many excellent articles that have appeared. Best wishes, Jon 
Cooper. 


Sent from ProtonMail mobile 



 Original Message  
On 3 Dec 2020, 11:17, Isabel Garcia-Saez < isabel.gar...@ibs.fr> wrote: 





Dear all, 

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix). 

[ https://www.nature.com/articles/d41586-020-03348-4 | 
https://www.nature.com/articles/d41586-020-03348-4 ] 

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times. 

Cheers, 

Isabel 


Isabel Garcia-Saez PhD 
Institut de Biologie Structurale 
Viral Infection and Cancer Group (VIC)-Cell Division Team 
71, Avenue des Martyrs 
CS 10090 
38044 Grenoble Cedex 9 
France 
Tel.: 00 33 (0) 457 42 86 15 
[ mailto:isabel.gar...@ibs.fr | e-mail: isabel.gar...@ibs.fr ] 
FAX: 00 33 (0) 476 50 18 90 
http://www.ibs.fr/ 





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



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



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[David Briggs]

A quick note regarding the code that Deepmind released for CASP13 (2018).

It bears the rather important caveat that: "This code can't be used to predict 
structure of an arbitrary protein sequence. It can be used to predict structure 
only on the CASP13 dataset (links below)."

Source: 
https://github.com/deepmind/deepmind-research/tree/master/alphafold_casp13

So whilst we can replicate their previous efforts, we currently can't submit 
our more troublesome sequences to their software, which is (I imagine) 
something that many of us might like to try.

D

--
Dr David C. Briggs
Senior Laboratory Research Scientist
Signalling and Structural Biology Lab
The Francis Crick Institute
London, UK
==
about.me/david_briggs


From: CCP4 bulletin board  on behalf of Isabel 
Garcia-Saez 
Sent: Thursday, December 3, 2020, 11:17
To: CCP4BB@JISCMAIL.AC.UK
Subject: [ccp4bb] AlphaFold: more thinking and less pipetting (?)

Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4

Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times.

Cheers,

Isabel


Isabel Garcia-Saez PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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


The Francis Crick Institute Limited is a registered charity in England and 
Wales no. 1140062 and a company registered in England and Wales no. 06885462, 
with its registered office at 1 Midland Road London NW1 1AT



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Jon Cooper]

Hello. A quick look suggests that a lot of the test structures were solved by 
phaser or molrep, suggesting it is a very welcome improvement on homology 
modelling. It would be interesting to know how it performs with structures of 
new or uncertain fold, if there are any left these days. Without resorting to 
jokes about artificial intelligence, I couldn't make that out from the CASP14 
website or the many excellent articles that have appeared. Best wishes, Jon 
Cooper.

Sent from ProtonMail mobile

 Original Message 
On 3 Dec 2020, 11:17, Isabel Garcia-Saez wrote:

> Dear all,
>
> Just commenting that after the stunning performance of AlphaFold that uses AI 
> from Google maybe some of us we could dedicate ourselves to the noble art of 
> gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
> everything together (just in case I have already prepared my subscription to 
> Netflix).
>
> https://www.nature.com/articles/d41586-020-03348-4
>
> Well, I suppose that we still have the structures of complexes (at the 
> moment). I am wondering how the labs will have access to this technology in 
> the future (would it be for free coming from the company DeepMind - Google?). 
> It seems that they have already published some code. Well, exciting times.
>
> Cheers,
>
> Isabel
>
> Isabel Garcia-SaezPhD
> Institut de Biologie Structurale
> Viral Infection and Cancer Group (VIC)-Cell Division Team
> 71, Avenue des Martyrs
> CS 10090
> 38044 Grenoble Cedex 9
> France
> Tel.: 00 33 (0) 457 42 86 15
> [e-mail: isabel.gar...@ibs.fr](mailto:isabel.gar...@ibs.fr)
> FAX: 00 33 (0) 476 50 18 90
> http://www.ibs.fr/
>
> ---
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] A postdoctoral position in structural biology is available at UT Southwestern Medical Center at Dallas

[Lijing Su]

Job Summary:

A postdoctoral position in structural biology is available in the Center
for the Genetics of Host Defense at UT Southwestern Medical Center at
Dallas. Using forward genetics, the lab identifies many proteins and
protein complexes that have novel functions in the immune system,
metabolism, and the nervous system. Many of these proteins and protein
complexes are involved in diseases including immune deficiency,
autoimmunity, inflammatory bowel disease, and diabetes. We are interested
in using structural biology to develop molecular mechanisms that describe
how these proteins and protein complexes function in the context of live
cells and mice.  UTSW-Medical Center is well equipped with state of the art
instrumentation for protein biochemistry, biophysics and structural
biology. This includes monthly access the Advanced Photo Source synchrotron
beamline, in-house robotic systems for screening and optimizing crystals,
various high strength NMR magnets equipped for solution state and solid
state NMR, and cryo-electron microscopes supported by the Cryo-Electron
Microscopy Facility at UTSW-Medical Center. New Ph.D. graduates are highly
encouraged to apply. Please visit the lab website at http://beutlerlab.org/ for
more information.



Successful candidates are expected to be highly motivated and well trained
in protein biochemistry, or in structural biology using NMR, X-ray
crystallography or cryoEM. Please send the following to Betsy Layton at
betsy.lay...@utsouthwestern.edu, with the subject line NEW Postdoctoral
Application in Structural Biology:

1)   A cover letter explaining your current research interests with a
description of why you are interested in the Beutler laboratory.

2)   A complete CV with links to all of your publications.

3)   A list of three references, with their current work address, title,
email address, and phone number.



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] Coming July 29: Improved Carbohydrate Data at the PDB -- N-glycans are now separate chains if more than one residue

[Robbie Joosten]

Dear Zhijie,

In generally I like the treatment of carbohydrates now as branched polymers. I 
didn't realise there was an exception. It makes sense for unlinked carbohydrate 
ligands, but not for N- or O-glycosylation sites as these might change during 
model building or, in my case, carbohydrate rebuilding in PDB-REDO powered by 
Coot. Thanks for pointing this out.

Cheers,
Robbie

> -Original Message-
> From: CCP4 bulletin board  On Behalf Of Zhijie Li
> Sent: Thursday, December 3, 2020 19:52
> To: CCP4BB@JISCMAIL.AC.UK
> Subject: Re: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the
> PDB -- N-glycans are now separate chains if more than one residue
> 
> Hi all,
> 
> I was confused when I saw mysterious new glycan chains emerging during
> PDB deposition and spent quite some time trying to find out what was
> wrong with my coordinates.  Then it occurred to me that a lot of recent
> structures also had tens of N-glycan chains.  Finally I realized that this
> phenomenon is a consequence of this PDB policy announced here in July.
> 
> 
> For future depositors who might also get puzzled, let's put it in a short
> sentence:  O- and N-glycans are now separate chains if it they contain more
> than one residue; single residues remain with the protein chain.
> 
> 
> https://www.wwpdb.org/documentation/carbohydrate-remediation
> 
> "Oligosaccharide molecules are classified as a new entity type, branched,
> assigned a unique chain ID (_atom_site.auth_asym_id) and a new mmCIF
> category introduced to define the type of branching
> (_pdbx_entity_branch.type) . "
> 
> 
> 
> 
> 
> I found the differential treatment of single-residue glycans and multi-residue
> glycans not only bit lack of aesthetics but also misleading.  When a structure
> contains both NAG-NAG... and single NAG on N-glycosylation sites, it might
> be because of lack of density for building more residues, or because that
> some of the glycosylation sites are now indeed single NAGs (endoH etc.)
> while some others are not cleaved due to accessibility issues.Leaving NAGs
> on the protein chain while assigning NAG-NAG... to a new chain, feels like
> suggesting something about their true oligomeric state.
> 
> 
> For example, for cryoEM structures, when one only builds a single NAG at a
> site does not necessarily mean that the protein was treated by endoH. In
> fact all sites are extended to at least tri-Man in most cases. Then why
> keeping some sites associated with the protein chain while others kicked
> out?
> 
> Zhijie
> 
> 
> 
> 
> 
> From: CCP4 bulletin board  on behalf of John
> Berrisford 
> Sent: Thursday, July 9, 2020 4:39 AM
> To: CCP4BB@JISCMAIL.AC.UK 
> Subject: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB
> 
> 
> Dear CCP4BB
> 
> PDB data will shortly incorporate a new data representation for
> carbohydrates in PDB entries and reference data that improves the
> Findability and Interoperability of these molecules in macromolecular
> structures. In order to remediate and improve the representation of
> carbohydrates across the archive, the wwPDB has:
> 
> * standardized Chemical Component Dictionary nomenclature
> following IUPAC-IUBMB recommendations
> * provided uniform representation for oligosaccharides
> * adopted Glycoscience-community commonly used linear descriptors
> using community tools
> * annotated glycosylation sites in PDB structures
> 
> Starting July 29, 2020, users will be able to access the improved data via FTP
> or wwPDB partner websites. Detailed information about this project is
> available at the wwPDB website
>  ; lists
> of impacted entries and chemical components will be published on this page
> after data release.
> 
> The wwPDB has created a new ‘branched’ entity representation for
> polysaccharides, describing all the individual monosaccharide components of
> these in the PDB entry. As part of this process, we have standardized atom
> nomenclature of >1,000 monosaccharides in the Chemical Component
> Dictionary (CCD) and applied a branched entity representation to
> oligosaccharides for >8000 PDB entries. To guarantee unambiguous chemical
> description of oligosaccharides in the affected PDB entries, an explicit
> description of covalent linkage information between their monosaccharide
> units is included. In addition, wwPDB validation reports provide consistent
> representation for these oligosaccharides and include 2D representations
> based on the Symbol Nomenclature for Glycans (SNFG).
> 
> To support the remediation of carbohydrate representation, software tools
> providing linear descriptors were developed in collaboration with the
> glycoscience community to enable easy translation of PDB data to other
> representations commonly used by glycobiologists. These include Condense
> IUPAC from GMML   at University
> of Georgia, 

Re: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB -- N-glycans are now separate chains if more than one residue

[Zhijie Li]

Hi all,

I was confused when I saw mysterious new glycan chains emerging during PDB 
deposition and spent quite some time trying to find out what was wrong with my 
coordinates.  Then it occurred to me that a lot of recent structures also had 
tens of N-glycan chains.  Finally I realized that this phenomenon is a 
consequence of this PDB policy announced here in July.

For future depositors who might also get puzzled, let's put it in a short 
sentence:  O- and N-glycans are now separate chains if it they contain more 
than one residue; single residues remain with the protein chain.

https://www.wwpdb.org/documentation/carbohydrate-remediation
"Oligosaccharide molecules are classified as a new entity type, branched, 
assigned a unique chain ID (_atom_site.auth_asym_id) and a new mmCIF category 
introduced to define the type of branching (_pdbx_entity_branch.type) . "


I found the differential treatment of single-residue glycans and multi-residue 
glycans not only bit lack of aesthetics but also misleading.  When a structure 
contains both NAG-NAG... and single NAG on N-glycosylation sites, it might be 
because of lack of density for building more residues, or because that some of 
the glycosylation sites are now indeed single NAGs (endoH etc.) while some 
others are not cleaved due to accessibility issues.Leaving NAGs on the 
protein chain while assigning NAG-NAG... to a new chain, feels like suggesting 
something about their true oligomeric state.

For example, for cryoEM structures, when one only builds a single NAG at a site 
does not necessarily mean that the protein was treated by endoH. In fact all 
sites are extended to at least tri-Man in most cases. Then why keeping some 
sites associated with the protein chain while others kicked out?

Zhijie




From: CCP4 bulletin board  on behalf of John Berrisford 

Sent: Thursday, July 9, 2020 4:39 AM
To: CCP4BB@JISCMAIL.AC.UK 
Subject: [ccp4bb] Coming July 29: Improved Carbohydrate Data at the PDB


Dear CCP4BB

PDB data will shortly incorporate a new data representation for carbohydrates 
in PDB entries and reference data that improves the Findability and 
Interoperability of these molecules in macromolecular structures. In order to 
remediate and improve the representation of carbohydrates across the archive, 
the wwPDB has:

  *   standardized Chemical Component Dictionary nomenclature following 
IUPAC-IUBMB recommendations
  *   provided uniform representation for oligosaccharides
  *   adopted Glycoscience-community commonly used linear descriptors using 
community tools
  *   annotated glycosylation sites in PDB structures

Starting July 29, 2020, users will be able to access the improved data via FTP 
or wwPDB partner websites. Detailed information about this project is available 
at the wwPDB 
website; lists of 
impacted entries and chemical components will be published on this page after 
data release.

The wwPDB has created a new ‘branched’ entity representation for 
polysaccharides, describing all the individual monosaccharide components of 
these in the PDB entry. As part of this process, we have standardized atom 
nomenclature of >1,000 monosaccharides in the Chemical Component Dictionary 
(CCD) and applied a branched entity representation to oligosaccharides for 
>8000 PDB entries. To guarantee unambiguous chemical description of 
oligosaccharides in the affected PDB entries, an explicit description of 
covalent linkage information between their monosaccharide units is included. In 
addition, wwPDB validation reports provide consistent representation for these 
oligosaccharides and include 2D representations based on the Symbol 
Nomenclature for Glycans (SNFG).

To support the remediation of carbohydrate representation, software tools 
providing linear descriptors were developed in collaboration with the 
glycoscience community to enable easy translation of PDB data to other 
representations commonly used by glycobiologists. These include Condense IUPAC 
from GMML at University of Georgia, 
WURCS from PDB2Glycan at The Noguchi 
Institute, Japan, and LINUCS 
from pdb-care at Germany.

Furthermore, to ensure continued Findability of 118 common oligosaccharides 
(e.g., sucrose, Lewis Y antigen), we have expanded the Biologically Interesting 
molecule Reference Dictionary (BIRD) that 
contains the covalent linkage information and common synonyms for such 
molecules.

wwPDB has also used this opportunity to improve the organization of chemical 
synonyms in the CCD by introducing a new _pdbx_chem_comp_synonyms data 
category. This will enable more comprehensive capture of alternative names for 
small molecules in the PDB. To minimize disruption to users, the legacy data 
item, 

Re: [ccp4bb] [phenixbb] [ccp4bb] Phenix refine distorting a sidechain despite correct density

[Igor Petrik]

 Thanks for the advice Pavel!

I don't want to celebrate prematurely, but I just re-ran the same
refinement in both the stable and in nightly 4070, and it _seems_ to be
corrected in the nightly. Will let you know if it pops up again.

- Igor Petrik, PhD


On Thu, Dec 3, 2020 at 9:16 AM Pavel Afonine  wrote:

> Hi Igor,
>
> please make sure to try the same refinement using the latest Phenix
> version from nightly builds first:
>
> http://phenix-online.org/download/nightly_builds.cgi
>
> The latest I see here is 1.19rc7-4070.
>
> Chances are the version you used had a bug that we fixed by now.
>
> If the problem persists please go ahead and send me inputs and indicate
> residues that are bad and I will investigate right away!
>
> Pavel
>
> On 12/3/20 08:42, Igor Petrik wrote:
>
> Thanks Nigel,
>
> I will put together the input and output file from my latest refinement
> and send them to you and Pavel later today.
>
> - Igor Petrik, PhD
>
>
> On Wed, Dec 2, 2020 at 9:46 PM Folmer Fredslund  wrote:
>
>> Dear Igor
>>
>> There's a phenix bulletin board for question like this, which is where
>> you should post the question. ( I crosspost here)
>>
>>
>> I would choose to not do the real space refinement in phenix.refine
>> during the last rounds of refinement of a model, when sidechain positions
>> are essentially correct.
>>
>>
>> I hope this helps
>>
>> Folmer
>>
>>
>>
>> tor. 3. dec. 2020 05.48 skrev Igor Petrik :
>>
>>> I am refining a 1.71A X-ray structure with phenix refine. I have
>>> everything modelled in - ~150 residues in the ASU and a heme - and my
>>> R-work/R-free is 0.17/0.22. But when I went to deposit it, PDB pointed out
>>> that two of my sidechains have distorted geometries. One is a His, and
>>> looking at it in Coot, I can clearly see the 2Fo-Fc density for the correct
>>> geometry, but the actual coordinates that phenix refine produce don't lie
>>> in that density; there are significant difference map peaks showing that
>>> the coordinates are in the wrong place. If I use real space refine in Coot
>>> to put the coordinates back into the correct density and refine it again in
>>> phenix, they get distorted again.
>>>
>>> What settings in phenix should I check to try to get it to properly
>>> refine the coordinates?
>>>
>>> Thanks,
>>> - Igor Petrik, PhD
>>>
>>> --
>>>
>>> To unsubscribe from the CCP4BB list, click the following link:
>>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
>>>
>>
> ___
> phenixbb mailing 
> listphenixbb@phenix-online.orghttp://phenix-online.org/mailman/listinfo/phenixbb
> Unsubscribe: phenixbb-le...@phenix-online.org
>
>
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] Fully-funded PhD studentships - Nottingham Trent University - UK

[Campeotto, Ivan]

Dear Colleagues,

I would like to bring to your attention the Fully-funded PhD studentships 
sponsored by the Doctoral School at Nottingham Trent University  (deadline 4th 
Jan 2021).

The PhD Programme is open to UK and non-UK applicants and information about the 
eligibility criteria and the program can be found at these links:

https://www.ntu.ac.uk/research/research-degrees-at-ntu/phd-studentships

https://www.ntu.ac.uk/research/research-degrees-at-ntu/phd-studentships/apply-for-a-phd-studentship

https://ntustudentship.smapply.io/prog/2021_nottingham_trent_university_phd_studentship_scheme/

The candidates need to contact the group leader directly before submitting a 
short research proposal as part of the application package.

If you know any potential candidate with interest in protein crystallography 
applied to Chagas disease and host inflammatory response for drug design and 
immunogen design applications, please suggest to contact me directly and send 
me a CV to discuss this further before the Christmas break. The project is in 
collaboration with the London School of Hygiene and Tropical Medicine and with 
the Medical University of Vienna.

More details about my research are available at 
https://www.ntu.ac.uk/staff-profiles/science-technology/ivan-campeotto


Best Wishes,

Dr Ivan Campeotto
Senior Lecturer in Biochemistry
Interdisciplinary Biomedical Research Centre - Room 115
School of Science and Technology
Nottingham Trent University
Clifton Campus - Nottingham - NG1 8NS
tel: +44 (0) 115 8483459



DISCLAIMER: This email is intended solely for the addressee. It may contain 
private and confidential information. If you are not the intended addressee, 
please take no action based on it nor show a copy to anyone. In this case, 
please reply to this email to highlight the error. Opinions and information in 
this email that do not relate to the official business of Nottingham Trent 
University shall be understood as neither given nor endorsed by the University. 
Nottingham Trent University has taken steps to ensure that this email and any 
attachments are virus-free, but we do advise that the recipient should check 
that the email and its attachments are actually virus free. This is in keeping 
with good computing practice.



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] Phenix refine distorting a sidechain despite correct density

[Igor Petrik]

Thanks Nigel,

I will put together the input and output file from my latest refinement and
send them to you and Pavel later today.

- Igor Petrik, PhD


On Wed, Dec 2, 2020 at 9:46 PM Folmer Fredslund  wrote:

> Dear Igor
>
> There's a phenix bulletin board for question like this, which is where you
> should post the question. ( I crosspost here)
>
>
> I would choose to not do the real space refinement in phenix.refine during
> the last rounds of refinement of a model, when sidechain positions are
> essentially correct.
>
>
> I hope this helps
>
> Folmer
>
>
>
> tor. 3. dec. 2020 05.48 skrev Igor Petrik :
>
>> I am refining a 1.71A X-ray structure with phenix refine. I have
>> everything modelled in - ~150 residues in the ASU and a heme - and my
>> R-work/R-free is 0.17/0.22. But when I went to deposit it, PDB pointed out
>> that two of my sidechains have distorted geometries. One is a His, and
>> looking at it in Coot, I can clearly see the 2Fo-Fc density for the correct
>> geometry, but the actual coordinates that phenix refine produce don't lie
>> in that density; there are significant difference map peaks showing that
>> the coordinates are in the wrong place. If I use real space refine in Coot
>> to put the coordinates back into the correct density and refine it again in
>> phenix, they get distorted again.
>>
>> What settings in phenix should I check to try to get it to properly
>> refine the coordinates?
>>
>> Thanks,
>> - Igor Petrik, PhD
>>
>> --
>>
>> To unsubscribe from the CCP4BB list, click the following link:
>> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
>>
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] cannot access ccp4 download webpage

[Johannes Cramer]

Dear all,

seems like one just needs to complain about it... It works again ;)

Thanks to all who replied!

Cheers,
Johannes

Am Do., 3. Dez. 2020 um 15:54 Uhr schrieb Jon Cooper <
jon.b.coo...@protonmail.com>:

> Works for me. Cheers, Jon.C.
>
>
> Sent from ProtonMail mobile
>
>
>
>  Original Message 
> On 3 Dec 2020, 14:47, Johannes Cramer < johannes.cra...@gmail.com> wrote:
>
>
> Dear all,
>
> I cannot access the ccp4 downloads webpage from my institute's pc or
> mobile.
> Is it just me? Is it a known problem? I get a 502 Proxy Error.
>
> http://www.ccp4.ac.uk/download/
>
> Cheers,
> Johannes
>
> --
>
> To unsubscribe from the CCP4BB list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=1
>
>



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] cannot access ccp4 download webpage

[Johannes Cramer]

Dear all,

I cannot access the ccp4 downloads webpage from my institute's pc or
mobile.
Is it just me? Is it a known problem? I get a 502 Proxy Error.

http://www.ccp4.ac.uk/download/

Cheers,
Johannes



To unsubscribe from the CCP4BB list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=CCP4BB=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] AlphaFold: more thinking and less pipetting (?)

[Isabel Garcia-Saez]

Dear all,

Just commenting that after the stunning performance of AlphaFold that uses AI 
from Google maybe some of us we could dedicate ourselves to the noble art of 
gardening, baking, doing Chinese Calligraphy, enjoying the clouds pass or 
everything together (just in case I have already prepared my subscription to 
Netflix).

https://www.nature.com/articles/d41586-020-03348-4 


Well, I suppose that we still have the structures of complexes (at the moment). 
I am wondering how the labs will have access to this technology in the future 
(would it be for free coming from the company DeepMind - Google?). It seems 
that they have already published some code. Well, exciting times. 

Cheers,

Isabel


Isabel Garcia-Saez  PhD
Institut de Biologie Structurale
Viral Infection and Cancer Group (VIC)-Cell Division Team
71, Avenue des Martyrs
CS 10090
38044 Grenoble Cedex 9
France
Tel.: 00 33 (0) 457 42 86 15
e-mail: isabel.gar...@ibs.fr
FAX: 00 33 (0) 476 50 18 90
http://www.ibs.fr/




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


smime.p7s
Description: S/MIME cryptographic signature

Re: [ccp4bb] phenix.refine with ligand with ambiguous electron density

[Dirk Kostrewa]

A way to achieve a feeling for the significance of the different map 
features that you are looking at in your current project, is the old 
control technique to omit both a well ordered part of the model 
(residue, alpha-helical turn, part of a beta-strand, ...) and a less 
well ordered part (residue, loop, end of visible secondary structure, 
...). Then you run refinement until convergence and calculate the 
different maps that you want to try. If you now look at the appearance 
of these maps at different contour levels for both the well and less 
well ordered omitted parts, and compare this with the appearance of the 
same maps at the same contour levels for your questionable part (bound 
ligand?), you can quickly get a feeling, how far you can trust what you 
see in these maps.


I learned this at the beginning of my crystallographic work and still do 
it today if I'm in doubt.


Cheers,

Dirk.

On 03.12.20 10:20, Robert Nicholls wrote:

Hi Dale,

You're absolutely right - the multiple hypothesis testing problem is 
one that is often not considered, let alone properly accounted for. 
Whilst this can be accounted for by appropriate adjustment of 
significance levels when a known number of explicit hypotheses are 
tested (and when estimated sigmas are appropriate and reliable...), 
this is extremely difficult in the present context when we passively 
conduct a large number of quick map evaluations subjectively by eye. 
Objective guidelines in such a case, which don't essentially boil down 
to an automated procedure, or unduly inhibit the process in other 
ways, would be valuable. I don't think there's a clear answer to this 
today, although raising awareness of such issues is very prudent. 
Indeed, there is an outstanding need for additional approaches for 
cross-validation, and perhaps re-evaluation of policies 
regarding provision of evidence of the reproducibility of 
crystallographic models. You're correct to say that, ultimately, there 
is (presently) no substitute for education and experience.


Best regards,
Rob


On 3 Dec 2020, at 08:09, Dale Tronrud > wrote:


Hi,

  Dr Nicholls brings up many interesting points, but doesn't touch on 
the major point I had hoped to make in my letter.  Whenever you start 
making multiple tests of your hypothesis you have to evaluate each of 
those tests with a higher standard than you would if you only applied 
one.  If you take a survey of the amount of fat people eat along with 
their history of heart disease you can calculate a correlation and 
find it significant with a p value of 0.05.  If, instead, you perform 
a survey asking for twenty different dietary behaviors and twenty 
health outcomes and find a correlation between eating fat and heart 
disease you need a much higher "signal" to determine its 
significance.  You just made 400 comparisons and a p of 0.05 allows 
20 spurious correlations to appear significant.


  If you are exploring your data set to decide if a compound has 
bound, and your try several different refinement programs and 
calculate several different map types based on the results of those 
refinements, and then adjust the blur of each map, and pick the map 
with the strongest peak in the putative binding site, you have to 
consider the significance of that peak height to be less than if you 
had just calculated one map and got that same height.


  Ignoring this counterintuitive fact has resulted in a huge number 
of studies in many fields to be published that ultimately turned out 
to not be reproducible.  It likely has also resulted in the 
deposition of a lot of "complex" models in the PDB that aren't correct.


  Yes, I am arguing for an ideal, hoping to pull some of you over 
toward my side a bit.  I certainly understand that one has to be 
flexible when solving a difficult problem, but you can't ignore that 
this "flexibility" has significant consequences for understanding the 
results of your work.


  Dr Nicholls' letter brings up a related topic which I'd like to 
explore.  His letter repeatedly mentions the importance of 
"intuition" when interpreting a map.  Yes, the power of human 
intuition, and our inability to replicate it in silico is the reason 
we are still staring at maps in Coot.  Intuition is a remarkable tool 
which, by its nature, is difficult to describe.


  Yet, no one is born with an innate intuition for interpreting 
electron density maps.  Intuition is acquired thru practice. 
 Practice is not simple repetition, however.  You can't become 
proficient in shooting basketball hoops by simply repeatedly throwing 
a basketball on the roof of your garage.  You have to have a proper 
backboard and a hoop.  Now, after repeatedly throwing the ball and 
"feeling" the difference between it going through the hoop and not, 
you will develop the ability to make a basket w/o really thinking 
about it.  You will have developed an intuition for achieving that task.


  There are two caveats.  

Re: [ccp4bb] phenix.refine with ligand with ambiguous electron density

[Robert Nicholls]

Hi Dale,

You're absolutely right - the multiple hypothesis testing problem is one that 
is often not considered, let alone properly accounted for. Whilst this can be 
accounted for by appropriate adjustment of significance levels when a known 
number of explicit hypotheses are tested (and when estimated sigmas are 
appropriate and reliable...), this is extremely difficult in the present 
context when we passively conduct a large number of quick map evaluations 
subjectively by eye. Objective guidelines in such a case, which don't 
essentially boil down to an automated procedure, or unduly inhibit the process 
in other ways, would be valuable. I don't think there's a clear answer to this 
today, although raising awareness of such issues is very prudent. Indeed, there 
is an outstanding need for additional approaches for cross-validation, and 
perhaps re-evaluation of policies regarding provision of evidence of the 
reproducibility of crystallographic models. You're correct to say that, 
ultimately, there is (presently) no substitute for education and experience.

Best regards,
Rob


> On 3 Dec 2020, at 08:09, Dale Tronrud  wrote:
> 
> Hi,
> 
>   Dr Nicholls brings up many interesting points, but doesn't touch on the 
> major point I had hoped to make in my letter.  Whenever you start making 
> multiple tests of your hypothesis you have to evaluate each of those tests 
> with a higher standard than you would if you only applied one.  If you take a 
> survey of the amount of fat people eat along with their history of heart 
> disease you can calculate a correlation and find it significant with a p 
> value of 0.05.  If, instead, you perform a survey asking for twenty different 
> dietary behaviors and twenty health outcomes and find a correlation between 
> eating fat and heart disease you need a much higher "signal" to determine its 
> significance.  You just made 400 comparisons and a p of 0.05 allows 20 
> spurious correlations to appear significant.
> 
>   If you are exploring your data set to decide if a compound has bound, and 
> your try several different refinement programs and calculate several 
> different map types based on the results of those refinements, and then 
> adjust the blur of each map, and pick the map with the strongest peak in the 
> putative binding site, you have to consider the significance of that peak 
> height to be less than if you had just calculated one map and got that same 
> height.
> 
>   Ignoring this counterintuitive fact has resulted in a huge number of 
> studies in many fields to be published that ultimately turned out to not be 
> reproducible.  It likely has also resulted in the deposition of a lot of 
> "complex" models in the PDB that aren't correct.
> 
>   Yes, I am arguing for an ideal, hoping to pull some of you over toward my 
> side a bit.  I certainly understand that one has to be flexible when solving 
> a difficult problem, but you can't ignore that this "flexibility" has 
> significant consequences for understanding the results of your work.
> 
>   Dr Nicholls' letter brings up a related topic which I'd like to explore.  
> His letter repeatedly mentions the importance of "intuition" when 
> interpreting a map.  Yes, the power of human intuition, and our inability to 
> replicate it in silico is the reason we are still staring at maps in Coot.  
> Intuition is a remarkable tool which, by its nature, is difficult to describe.
> 
>   Yet, no one is born with an innate intuition for interpreting electron 
> density maps.  Intuition is acquired thru practice.  Practice is not simple 
> repetition, however.  You can't become proficient in shooting basketball 
> hoops by simply repeatedly throwing a basketball on the roof of your garage.  
> You have to have a proper backboard and a hoop.  Now, after repeatedly 
> throwing the ball and "feeling" the difference between it going through the 
> hoop and not, you will develop the ability to make a basket w/o really 
> thinking about it.  You will have developed an intuition for achieving that 
> task.
> 
>   There are two caveats.  First, you have to actually watch the ball go 
> through the hoop.  If you close your eyes right after your throw you will 
> never develop a useful skill.  It is the feedback from the success or failure 
> of each attempt that makes it practice.  Second, no matter how much time you 
> spend shooting baskets, you will never get better at dribbling the ball.  
> Good practice allows you to develop intuition, but only intuition about that 
> task.
> 
>   Let's say you are working on a project, but having difficulty interpreting 
> your map at some critical location.  You ask around and learn of some spiffy 
> new map calculation and you want to try it.  While you certainly can 
> calculate the map, you have no intuition on how to interpret it.  You have 
> not practiced with that type of map.
> 
>   It may look similar to the maps you've looked at before, but that 
> similarity can be a 

Re: [ccp4bb] phenix.refine with ligand with ambiguous electron density

[Dale Tronrud]

Hi,

   Dr Nicholls brings up many interesting points, but doesn't touch on 
the major point I had hoped to make in my letter.  Whenever you start 
making multiple tests of your hypothesis you have to evaluate each of 
those tests with a higher standard than you would if you only applied 
one.  If you take a survey of the amount of fat people eat along with 
their history of heart disease you can calculate a correlation and find 
it significant with a p value of 0.05.  If, instead, you perform a 
survey asking for twenty different dietary behaviors and twenty health 
outcomes and find a correlation between eating fat and heart disease you 
need a much higher "signal" to determine its significance.  You just 
made 400 comparisons and a p of 0.05 allows 20 spurious correlations to 
appear significant.


   If you are exploring your data set to decide if a compound has 
bound, and your try several different refinement programs and calculate 
several different map types based on the results of those refinements, 
and then adjust the blur of each map, and pick the map with the 
strongest peak in the putative binding site, you have to consider the 
significance of that peak height to be less than if you had just 
calculated one map and got that same height.


   Ignoring this counterintuitive fact has resulted in a huge number of 
studies in many fields to be published that ultimately turned out to not 
be reproducible.  It likely has also resulted in the deposition of a lot 
of "complex" models in the PDB that aren't correct.


   Yes, I am arguing for an ideal, hoping to pull some of you over 
toward my side a bit.  I certainly understand that one has to be 
flexible when solving a difficult problem, but you can't ignore that 
this "flexibility" has significant consequences for understanding the 
results of your work.


   Dr Nicholls' letter brings up a related topic which I'd like to 
explore.  His letter repeatedly mentions the importance of "intuition" 
when interpreting a map.  Yes, the power of human intuition, and our 
inability to replicate it in silico is the reason we are still staring 
at maps in Coot.  Intuition is a remarkable tool which, by its nature, 
is difficult to describe.


   Yet, no one is born with an innate intuition for interpreting 
electron density maps.  Intuition is acquired thru practice.  Practice 
is not simple repetition, however.  You can't become proficient in 
shooting basketball hoops by simply repeatedly throwing a basketball on 
the roof of your garage.  You have to have a proper backboard and a 
hoop.  Now, after repeatedly throwing the ball and "feeling" the 
difference between it going through the hoop and not, you will develop 
the ability to make a basket w/o really thinking about it.  You will 
have developed an intuition for achieving that task.


   There are two caveats.  First, you have to actually watch the ball 
go through the hoop.  If you close your eyes right after your throw you 
will never develop a useful skill.  It is the feedback from the success 
or failure of each attempt that makes it practice.  Second, no matter 
how much time you spend shooting baskets, you will never get better at 
dribbling the ball.  Good practice allows you to develop intuition, but 
only intuition about that task.


   Let's say you are working on a project, but having difficulty 
interpreting your map at some critical location.  You ask around and 
learn of some spiffy new map calculation and you want to try it.  While 
you certainly can calculate the map, you have no intuition on how to 
interpret it.  You have not practiced with that type of map.


   It may look similar to the maps you've looked at before, but that 
similarity can be a trap.  By now a large number of us here on the BB 
have had the experience of looking at a high resolution electrostatic 
potential (ESP) map and "feeling" that something is wrong with it.  The 
carbonyl oxygen bumps are too small and the acid groups are oddly weak. 
 Wow, those magnesium ions really stand out -- Maybe they're potassium 
instead?  No, there is nothing wrong with the ESP map.  The fault is 
with our intuition which was based on many, many hours of looking at ED 
maps.  To interpret ESP maps you have to practice with a bunch of ESP 
maps first.


   You cannot develop intuition for the spiffy map calculated from your 
project's data since you don't know its correct interpretation -- It 
cannot give you feedback.  Before you calculate this map for your data 
you should calculate versions for many other *completed* projects and 
get a "feel" for what that kind of map shows under different 
circumstances.  Practice, practice, practice, then you will be ready to 
return to your little mystery and be able to apply your, newly acquired, 
intuition.


   Yes, I try new refinement programs - But first I run refinement with 
them on familiar proteins. Yes, I try new styles of map calculations - 
But first I calculate those maps for