Re: [AMBER] How to analyse the PCA results?

From: Markowska <markowska.kar.gmail.com>
Date: Thu, 26 Jul 2018 15:41:40 +0200

Another question: as I have the Project.dat file with eigenvectors how to
obtain the eigenvalues?

Best regards,
Karolina

2018-07-26 15:10 GMT+02:00 Karolina Mitusińska (Markowska) <
markowska.kar.gmail.com>:

> Hi Dan!
>
> I've finally managed to obtain the Project.dat file and run the 'filter'
> command. Thank you very much for help.
> How to properly interpret the Project.dat file? Are they just a collection
> of eigenvectors in each simulation frame?
>
> And second question: are those structures in filtered.pc.nc comes from
> the original trajectories? If so, can I somehow extract their numbers from
> the trajectory?
>
> Best regards,
> Karolina
>
> 2018-07-20 16:23 GMT+02:00 Daniel Roe <daniel.r.roe.gmail.com>:
>
>> On Fri, Jul 20, 2018 at 10:20 AM, Daniel Roe <daniel.r.roe.gmail.com>
>> wrote:
>> >>
>> >> And if I finally get the "distribution of components" plot, how would I
>> >> know, which conformation is represented by which part of the plot?
>>
>> This can be done using the 'filter' command and your 'projection'
>> data. After reading in your trajectory and projection data:
>>
>> trajin MyTraj.nc
>> readdata Project.dat
>>
>> the 'filter' command can be used to select regions of interest. Note
>> that when reading the projections back in, the data set index will
>> refer to the column (eigenvector 1 is in column 2):
>>
>> # Select regions between -30 to -25 for PC1 and 10 to 15 for PC2
>> filter Project.dat:2 min -30 max -25 Project.dat:3 min 10 max 15
>> trajout filtered.pc.nc
>>
>> -Dan
>>
>> >>
>> >> Thanks for your help!
>> >> Karolina Mitusińska
>> >> PhD student
>> >> _______________________________________________
>> >> 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
Received on Thu Jul 26 2018 - 07:00:01 PDT
Custom Search