This is mostly moot since 'reproducibility' means being able to reproduce someone's conclusions and not their specific results. Something that a lot of funding agencies don't seem to understand when they talk about reproducibility and making data available etc etc. But my reading of the text in the paper
>>>>>>> "380 ps initial MD run with consecutively weaker restraints on
>>>>>>> protein atoms and gradually heat up the system to 300K"
is that they likely just ran 20ps with restraint blah, then another 20ps with restrain blah*0.x then another 20ps with blahx0.y etc etc. I don't see anything in that wording which refers to the term 'steadily' or 'linearly' decreasing restraints. I'd look in the supplemental material to figure out what they actually did - if it is not there then that is a little sloppy in terms of specifying the methodology used etc.
At the end of the day though, who the hell cares. If you have to reproduce their protocol exactly to reproduce their conclusions then the conclusions are worthless.
All the best
Ross
>
> On 05/18/2015 04:13 PM, Juan Eiros Zamora wrote:
>> Hi all,
>>
>> I'm aware that even if I do exactly the same preparation of the system I
>> will get differences on the results, as that is the nature of MD, but
>> all in all the conclusions of the study should be the same.
>>
>> I would ask for the scripts but the issue is that they used NAMD instead
>> of AMBER. Is there a way to use NAMD input files with AMBER? Or am I
>> just being silly here?
>>
>> Anyway thanks for the suggestions, for the moment I think I'll just do
>> it sequentially.
>>
>> Cheers,
>>
>> Juan
>>
>> On 18/05/15 15:00, Vlad Cojocaru wrote:
>>> Sure ... Hannes is also right. Even if you apply the same protocol is
>>> unlikely you'll get exactly the same results ... However, if you are
>>> after reproducing the overall conclusions of a study, they should be
>>> protocol-independent provided a protocol does not introduce artefactual
>>> behavior. So, probably in this case the best bet is to get the original
>>> scripts for the published protocols from the authors.
>>>
>>> Hope this helps a bit
>>> Vlad
>>>
>>> On 05/18/2015 03:56 PM, Vlad Cojocaru wrote:
>>>> Dear Juan,
>>>>
>>>> Probably the best way to reproduce published data is to ask the authors
>>>> for the protocols ..
>>>>
>>>> Best
>>>> Vlad
>>>>
>>>>
>>>> On 05/18/2015 03:37 PM, Juan Eiros Zamora wrote:
>>>>> Hi Hannes,
>>>>>
>>>>> It's just that I'm trying to replicate the MD protocol of a paper, but
>>>>> the authors used NAMD 2.9, and they stated in their Materials and
>>>>> Methods that they heat up the system and change the constraints at the
>>>>> same time. So I'm trying to introduce the least changes as possible in
>>>>> order to obtain similar results to them.
>>>>>
>>>>> But if it's not possible in AMBER then I will just do it sequentially as
>>>>> you suggested.
>>>>>
>>>>> Thanks for your input,
>>>>>
>>>>> Juan
>>>>>
>>>>>
>>>>>
>>>>> On 18/05/15 13:13, Hannes Loeffler wrote:
>>>>>> Hi,
>>>>>>
>>>>>> I am not aware that you could change the restraint force during MD so I
>>>>>> think you would have to emulate that through several steps. Possibly
>>>>>> script this.
>>>>>>
>>>>>> Why is it important to you to do the heating and restraint release at
>>>>>> the same time? What's wrong with doing this step-wise?
>>>>>>
>>>>>> Cheers,
>>>>>> Hannes.
>>>>>>
>>>>>>
>>>>>> On Mon, 18 May 2015 13:06:56 +0100
>>>>>> Juan Eiros Zamora <j.eiros-zamora14.imperial.ac.uk> wrote:
>>>>>>
>>>>>>> Hi everyone,
>>>>>>>
>>>>>>> I am trying to replicate an MD protocol that has been done previously
>>>>>>> using NAMD 2.9 using AMBER.
>>>>>>>
>>>>>>> After the initial minimisation steps (water and ions, then backbone
>>>>>>> of protein and so on), I am trying to do the following:
>>>>>>>
>>>>>>> "380 ps initial MD run with consecutively weaker restraints on
>>>>>>> protein atoms and gradually heat up the system to 300K"
>>>>>>>
>>>>>>> Usually in my minimisation input file I keep the restraints constant
>>>>>>> like this:
>>>>>>>> Stage 1 - 1000 step minimisation of water and ions
>>>>>>>> &cntrl
>>>>>>>> imin=1, maxcyc=1000, ncyc=500,
>>>>>>>> cut=12.,ntb=1,ntr=1,
>>>>>>>> ntpr=100
>>>>>>>> /
>>>>>>>> Hold protein fixed
>>>>>>>> 500.0
>>>>>>>> RES 1 422
>>>>>>>> END
>>>>>>>> END
>>>>>>> And to heat up a system I do it like this (after the &cntrl section):
>>>>>>>
>>>>>>>> &wt type='TEMP0', istep1=0, istep2=50000,
>>>>>>>> value1=0.0, value2=300.0 /
>>>>>>>> &wt type='END' /
>>>>>>>> Hold protein fixed
>>>>>>>> 10.0
>>>>>>>> RES 1 422
>>>>>>>> END
>>>>>>>> END
>>>>>>> Does anyone have an idea of how could I change two variables at a
>>>>>>> time? I've looked into section 18.8 of the Manual, but I am not sure
>>>>>>> if I should change the BOND option or the REST one, or how to proceed
>>>>>>> to change one of the two as well as the TEMP0 option.
>>>>>>>
>>>>>>> Looking forward to hear any suggestion on this matter,
>>>>>>>
>>>>>>> Juan
>>>>>>>
>>>>>>>
>>>>>>> _______________________________________________
>>>>>>> AMBER mailing list
>>>>>>> AMBER.ambermd.org
>>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>> _______________________________________________
>>>>>> AMBER mailing list
>>>>>> AMBER.ambermd.org
>>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>> _______________________________________________
>>>>> AMBER mailing list
>>>>> AMBER.ambermd.org
>>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>>>
>>>>>
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>>
>
> --
> Dr. Vlad Cojocaru
> Computational Structural Biology Laboratory
> Department of Cell and Developmental Biology
> Max Planck Institute for Molecular Biomedicine
> Röntgenstrasse 20, 48149 Münster, Germany
> Tel: +49-251-70365-324; Fax: +49-251-70365-399
> Email: vlad.cojocaru[at]mpi-muenster.mpg.de
> http://www.mpi-muenster.mpg.de/43241/cojocaru
>
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon May 18 2015 - 09:30:02 PDT
