So according to the manual:
"If the series keyword is specified the time series for each hydrogen bond (1 for present, 0 for not present) will also be saved for subsequent analysis (e.g. with lifetime, see on page 725)"
In the example presented below I thought I put the series data into S1 and saved the same data as “series.dat”. Hopefully this is what happens. And according to this entry in the manual I should be able to perform subsequent analysis using this file, correct?
Now reading further regarding data I suspect I need to read this data into CPPTRAJ, so I think this is done using the “readdata” command. After playing around with the command for a while I thought I managed to produce something that loads the data as such:
readdata series.dat intcols 2 name S1
Which I think would load my “series” data into a dataset named S1. Then I should, in theory, be able to run the lifetime analysis on this dataset, though I cannot figure the syntax out. At this stage I assumed that
lifetime out lifetime.dat S1[solutehb]
Should produce the same result as the previously suggested input though I got some errors regarding datasets not being read.
Warning: 'S2[solutehb]' selects no data sets.
Error: No data sets selected.
Error: lifetime: Could not add data sets.
Error: Could not setup analysis [lifetime]
So I ended up with this:
readdata series.dat intcols 2 name S1
lifetime out lifetime.dat S1
runanalysis
Which executes, producing output. However, the output is just for “one” hydrogen bond.
#Set lifetime_00001 lifetime_00001[max] lifetime_00001[avg] lifetime_00001[frames] lifetime_00001[name]
1 2042 1243 16.6988 34099 S1[UU]
Where I would have expected something like
#Set lifetime_00688 lifetime_00688[max] lifetime_00688[avg] lifetime_00688[frames] lifetime_00688[name]
1 6 17 4.8333 29 XXX_224.O-YYY_56.O-H3
2 18 95 16.0556 289 XXX_225.O-YYY_56.O-H3
3 10 15 6.3000 63 XXX_259.O-YYY_56.O-H3
4 17 18 7.1765 122 XXX_261.O-YYY_56.O-H3
As the defined mask involves multiple molecules of each kind.
Looking at the “series” input file I am not super surprised as there are only two columns with frame vs 0/1 so where would that information come from. This makes me suspect that my input:
> hbond S1 series out series.dat \
Is not enough to save required data to reproduce the same output “after the fact”. It seems that every “pair" gets its own “dataset” when I run the analysis in one swoop. So I would need some different type of file storing multiple sets with unique names in a multi column file. So for my current issue I suspect I need to redo the entire analysis run to get te missing files.
For future reference, what could I use in my HBOND analyses that would save enough data to run the lifetime analyses at a later point in time and produce/reproduce the expected output?
Best regards
// Gustaf
> On 28 Aug 2020, at 14:11, Gustaf Olsson <gustaf.olsson.lnu.se> wrote:
>
> So, turns out that after running a large set of HBOND analyses using cpptraj I find myself with some missing lifetime analyses output files.
>
> I have the avgout files and they seem to be correct. I have the series data saved and it is present and has content. Now, I am just missing the actual lifetime analyses output files.
>
> Roughly my input:
>
> hbond S1 series out series.dat \
> donormask :X.O donorhmask :X.H \
> acceptormask :Y.O \
> avgout average.dat nointramol
> run
> runanalysis lifetime S1[solutehb] out lifetime.dat
>
> I am, according to above, missing the “lifetime.dat” files for some runs.
>
> My question is this, can I run the “runanalysis lifetime” part again using the saved S1 set (series.dat) somehow, not having to rerun the entire analysis. If so, what would the input for such analysis look like?
>
> Best regards
> // Gustaf
> _______________________________________________
> 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 Aug 31 2020 - 03:30:02 PDT
