Discussion group for the ARIA software

[Benjamin Bardiaux <bardiaux AT pasteur.fr>]

Hi Mark,

Did you also change the violation_tolerance parameter for every iteration using 
the values from Supp. Table S5 of Mareuil et. al., J Bio NMR 2015 ?

We will do our best to try and make it easier to choose this protocol in the 
ARIA project/GUI.

Best,

Benjamin

On 10/06/2020 05:12, Kelly, Mark wrote:
> Dear All,
>
> I wanted to run the protocol of Mareuil et. al., J Bio NMR 2015 in ARIA 2.3.2 
> (including that of Nilges et, al., Structure 2008 and earlier papers).
>
> I made the following changes to my ARIA project.xml file (at the bottom), but 
> when I run the calculation the structures improve in the first iterations, but 
> then in spite of there being apparently only a few violations the number of noes 
> used suddenly goes down by about 1/2?
>
> any ideas what might be wrong?
>
> Thanks,
>
> Fitted:
> ## Backbone (N,CA,C,O) : 1.06 +/- 0.22 A
> ## Backbone (N,CA,C,O) : 1.04 +/- 0.24 A
> ## Backbone (N,CA,C,O) : 0.92 +/- 0.16 A
> ## Backbone (N,CA,C,O) : 0.89 +/- 0.21 A
> ## Backbone (N,CA,C,O) : 0.83 +/- 0.21 A
> ## Backbone (N,CA,C,O) : 0.86 +/- 0.17 A
> ## Backbone (N,CA,C,O) : 1.59 +/- 0.32 A
> ## Backbone (N,CA,C,O) : 1.80 +/- 0.48 A
> ## Backbone (N,CA,C,O) : 1.78 +/- 0.25 A
> ## Backbone (N,CA,C,O) : 1.90 +/- 0.34 A
> ## Used during calculation: 1988
> ## Used during calculation: 1988
> ## Used during calculation: 1921
> ## Used during calculation: 1443
> ## Used during calculation: 1449
> ## Used during calculation: 1450
> ## Used during calculation: 595
> ## Used during calculation: 624
> ## Used during calculation: 634
> ## Violated:                0 (0.0)
> ## Violated:                0 (0.0)
> ## Violated:                4 (0.2)
> ## Violated:                17 (1.2)
> ## Violated:                6 (0.4)
> ## Violated:                3 (0.2)
> ## Violated:                60 (10.1)
> ## Violated:                43 (6.9)
> ## Violated:                39 (6.2)
>
>
>
> *Changes to ARIA project.xml file:*
>
> (1) the softening of the force field:
>
> <parameter_definition name="parallhdg5.3softbig.pro" filename="”/>
>
>
> (2) log harmonic potential, Bayesian weighting of distance restraints (?) and 
> increased number of steps (table S5 in the supplementary):
>
> <logharmonic_potential enabled="yes" use_auto_weight="yes" weight_unambig="25.0" 
> weight_ambig="25.0" weight_hbond="25.0"/>
>        </annealing_parameters>
>        <md_parameters dynamics="torsion" random_seed="12468759" 
> tad_temp_high="10000.0" tad_timestep_factor="9.0" cartesian_temp_high="2000.0" 
> cartesian_first_iteration="0" timestep="0.002" temp_cool1_final="1000.0" 
> temp_cool2_final="50.0" steps_high="20000" steps_refine="4000" 
> steps_cool1="30000" steps_cool2="30000"/>
>
>
> (3) a procedure that automatically adjusts the violation tolerance used in the 
> selection of active restraints (?):
>
> <violation_analysis sigma_mode="auto" violation_tolerance="0.1" 
> violation_threshold="0.5"/>
>
> We used distance restraints file from ccpn:
>
> <unambiguous_distance_restraints file="" format="ccpn" ccpn_id="18|5" 
> enabled="yes" add_to_network="no" calibrate="all_iterations_except_first" 
> run_network_anchoring="no" filter_contributions="yes"/>
> <ambiguous_distance_restraints file="" format="ccpn" ccpn_id="18|6" 
> enabled="yes" add_to_network="no" calibrate="all_iterations_except_first" 
> run_network_anchoring="no" filter_contributions="yes”/>
>
>
> Do we need to specify lower/upper bounds with the restraints?
>
> <lower_bound_correction value="0.0" enabled="no"/>
> <upper_bound_correction value="6.0" enabled="no"/>
>
>
>
>
>
>
>
>
> > On Jun 4, 2020, at 10:18 AM, Kelly, Mark <Mark.Kelly AT ucsf.edu 
> > <mailto:Mark.Kelly AT ucsf.edu>> wrote:
> >
> >
> > Hi Vicky/All,
> >
> > we have been running our calculations in ARIA vers 2.3.1. If we want to take 
> > advantage of the log harmonic potential should we use 2.3.2 or is this only 
> > important for the adaptive violation tolerance (below)? Would we need 
> > different parameters for using just the log harmonic potential in 2.31, than 
> > those from Mareuil et. al., 2015) for the adaptive violation tolerance 
> > incorporated in 2.3.2. The linux workstation James is using has 2.3.1 that 
> > comes with SBGrid. I have a 2.3.2 version on my Mac I can try.
> >
> > Thanks for all your help,
> >
> > Mark
> >
> >
> >       The latest version (2.3.2)
> >
> > The source code of version 2.3 is available as a tarball: aria2.3.tar.gz 
> > <https://urldefense.proofpoint.com/v2/url?u=http-3A__aria.pasteur.fr_archives_aria2.3.2.tar.gz&d=DwMGaQ&c=iORugZls2LlYyCAZRB3XLg&r=298wo5FTw8hoVM2Z8-GREvQZopmfLGcC8YyMqGWWpmw&m=ylym3D2tfYPSSGTR-dqOKNPR0CkS38Jx05ca3cwk4X0&s=qHozzh-zB6f9WIwHHU4etIB9GRxLgtVnrq-lW_pLIsg&e=>
> >
> >   * Version 2.3.2 updates version 2.31 and  offers several new features:
> >
> >       o compatibility with CCPNmr Analsyis 2.4
> >       o new PDF reports
> >       o improved Zinc ions patches
> >       o Constraints combination
> >       o Automatic superimposition of best structures and bundle RMSD
> >       o *Adaptive violation tolerance (cf Mareuil et al. J Bio NMR 015)*
> >       o various fixes/
> >         /
> >
> >   * ARIA version 2.3 requires the python module *numpy* (Numeric not 
> > supported)
> >
> >
> >     Past versions
> >
> >
> >       Version 2.3.1
> >
> > The source code of version 2.3 is available as a tarball: aria2.3.tar.gz 
> > <https://urldefense.proofpoint.com/v2/url?u=http-3A__aria.pasteur.fr_archives_aria2.3.1.tar.gz&d=DwMGaQ&c=iORugZls2LlYyCAZRB3XLg&r=298wo5FTw8hoVM2Z8-GREvQZopmfLGcC8YyMqGWWpmw&m=ylym3D2tfYPSSGTR-dqOKNPR0CkS38Jx05ca3cwk4X0&s=92BqyQFaDa94JaW0RW5wvF7MpnMXpjbYFnan014NksM&e=>
> >
> >  *
> >     Version 2.3 is based on the ARIA software library.
> >   * Version 2.3 is an extension of the version 2.2 of the program. It offers
> >     several new features:
> >
> >       o extension of support for symmetric oligomers (C3, C5 and D2 
> > symmetries)
> >       o compatibility with *CCPNmr version 2* (version 1 not supported)
> >       o improved import/export of CCPN data (eg. import of structure models,
> >         NMR Calculation runs in Analysis...)
> >       o *introduction of a log-harmonic potential for distance restraints
> >         (Nilges /et al./ Structure, 2008)*
> >       o support for coordinated *zinc* ions
> >       o automated Molprobilty /clashlist /analysis
> >       o /partial support for solid-state NMR through CCPN Distance constraint
> >         lists (see Note
> >         
> > <https://urldefense.proofpoint.com/v2/url?u=http-3A__aria.pasteur.fr_documentation_tutorials_aria-2Dand-2Dccpn&d=DwMGaQ&c=iORugZls2LlYyCAZRB3XLg&r=298wo5FTw8hoVM2Z8-GREvQZopmfLGcC8YyMqGWWpmw&m=ylym3D2tfYPSSGTR-dqOKNPR0CkS38Jx05ca3cwk4X0&s=HoVS4Ts4MXRPLmKdp7XZlOoP7Efp947r1QvuaD3LCzI&e=>)/
> >
> >   * ARIA version 2.3 requires the python module *numpy* (Numeric not 
> > supported)
> >
> > /NB : The log-harmonic potential requires to recompile CNS with the sources 
> > provided in the aria archive (See README or installation notes) 
> > <https://urldefense.proofpoint.com/v2/url?u=http-3A__aria.pasteur.fr_downloads_installation-2Dnotes&d=DwMGaQ&c=iORugZls2LlYyCAZRB3XLg&r=298wo5FTw8hoVM2Z8-GREvQZopmfLGcC8YyMqGWWpmw&m=ylym3D2tfYPSSGTR-dqOKNPR0CkS38Jx05ca3cwk4X0&s=AkyNVlkWK6pyCEhHSCSKtdk7uQfN9DZhDpoNVkXf9Vg&e=>/
> >
> >
> >
> > > On Jun 2, 2020, at 11:41 PM, Higman, Vicky (Dr.) <vad5 AT LEICESTER.AC.UK 
> > > <mailto:vad5 AT LEICESTER.AC.UK>> wrote:
> > >
> > > Hi Mark,
> > >
> > > try using the parameters suggested in table S5 in the supplementary info of 
> > > the Mareuil et al. 2015 paper. I have generally had good experiences with 
> > > that. Note in particular the number of SA high temp and cooling steps - these 
> > > are much larger than what is typically used.
> > >
> > > All the best,
> > >
> > > Vicky
> > > --------------------------------------------------------------------------------
> > > *From:*aria-discuss-request AT services.cnrs.fr 
> > > <mailto:aria-discuss-request AT services.cnrs.fr><aria-discuss-request AT services.cnrs.fr 
> > > <mailto:aria-discuss-request AT services.cnrs.fr>> on behalf of Kelly, Mark 
> > > <Mark.Kelly AT ucsf.edu <mailto:Mark.Kelly AT ucsf.edu>>
> > > *Sent:*02 June 2020 18:17
> > > *To:*aria-discuss AT services.cnrs.fr 
> > > <mailto:aria-discuss AT services.cnrs.fr><aria-discuss AT services.cnrs.fr 
> > > <mailto:aria-discuss AT services.cnrs.fr>>
> > > *Cc:*Benjamin Bardiaux <bardiaux AT pasteur.fr <mailto:bardiaux AT pasteur.fr>>; 
> > > Longbotham, James <James.Longbotham AT ucsf.edu <mailto:James.Longbotham AT ucsf.edu>>
> > > *Subject:*Re: [aria-discuss] Next step in calculations
> > > Hello All,
> > >
> > > we switched ‘on’ the log harmonic potential, but now our structures don’t 
> > > look very nice. I wonder if we did not set the parameters correctly or do we 
> > > have to switch something else off?
> > >
> > > Best,
> > >
> > > Mark
> > >
> > > <logharmonic_potential enabled="yes" use_auto_weight="yes" 
> > > weight_unambig="1.0" weight_ambig="1.0" weight_hbond="1.0"/>
> > >
> > >
> > >
> > >
> > >
> > >
> > > > On Apr 29, 2020, at 10:50 AM, Kelly, Mark <Mark.Kelly AT ucsf.edu 
> > > > <mailto:Mark.Kelly AT ucsf.edu>> wrote:
> > > >
> > > > Hello All,
> > > >
> > > > we tried to restart a run that had been successful before and get the error 
> > > > below - would anyone have a suggestion what might be wrong?
> > > >
> > > > hope all are safe and well,
> > > >
> > > > Mark
> > > >
> > > >
> > > >
> > > > > Begin forwarded message:
> > > > >
> > > > > *From:*"Longbotham, James" <James.Longbotham AT ucsf.edu 
> > > > > <mailto:James.Longbotham AT ucsf.edu>>
> > > > > *Subject:**RE: Next step in calculations*
> > > > > *Date:*April 29, 2020 at 10:37:57 AM PDT
> > > > > *To:*"Kelly, Mark" <Mark.Kelly AT ucsf.edu <mailto:Mark.Kelly AT ucsf.edu>>
> > > > >
> > > > > Ok so that last run I  described below failed. I thought it ould have 
> > > > > something to do with the merged restraint lists so I tried to run it again 
> > > > > with the restraints I used previously (before assignment) and it failed 
> > > > > still! See error below. Not sure what’s going on
> > > > > Exception in thread Thread-1:
> > > > > Traceback (most recent call last):
> > > > >   File 
> > > > > "/programs/x86_64-linux/ccpnmr/2.4.2/python2.7/lib/python2.7/threading.py", 
> > > > > line 810, in __bootstrap_inner
> > > > >     self.run()
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/JobManager.py", line 
> > > > > 193, in run
> > > > >     f(self)
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/JobManager.py", line 
> > > > > 331, in job_done
> > > > >     self.dispatch_job(host)
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/JobManager.py", line 
> > > > > 337, in dispatch_job
> > > > >     self.done()
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/JobManager.py", line 
> > > > > 384, in done
> > > > >     self.__callback()
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/cns.py", line 2662, 
> > > > > in <lambda>
> > > > >     callback = lambda v = vars, f = self.refine_done: f(v)
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/cns.py", line 2522, 
> > > > > in refine_done
> > > > >     self.__callback()
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/Protocol.py", line 
> > > > > 654, in <lambda>
> > > > >     c = self.structure_calc_done: c(engine, molecule, it)
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/Protocol.py", line 
> > > > > 624, in structure_calc_done
> > > > >     ensemble = engine.getEnsemble(it_settings, molecule)
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/cns.py", line 666, in 
> > > > > getEnsemble
> > > > >     ensemble.read(pdbfiles, molecule, float_files = swapped_atoms)
> > > > >   File 
> > > > > "/programs/x86_64-linux/aria/2.3.1/src/py/aria/StructureEnsemble.py", line 
> > > > > 194, in read
> > > > >     pdb_dict = self.reader.read(files[j], chain_types, format[j])
> > > > >   File "/programs/x86_64-linux/aria/2.3.1/src/py/aria/PDBReader.py", line 
> > > > > 186, in read
> > > > >    dict['noe_violations'] = float(data[13:].split(',')[0])
> > > > > ValueError: could not convert string to float: D1_ApoH3_violations_1.pdb"


--
---------------------------------------------
Dr Benjamin Bardiaux      bardiaux AT pasteur.fr
Unité de Bioinformatique Structurale
CNRS UMR3528 - Institut Pasteur
25,28 rue du Docteur Roux 75015 Paris, France
---------------------------------------------