I would like to calculate auto-correlation function for tilt angle for the lipids in a bilayer from molecular dynamic simulation. I have the angles for all lipids from each frames for all subsequent frames but I need a code to calculate the auto-correlation function. Where can I find or get the code apart from writing my own?
Autocorrelation function is implemented in many (commercial/open) software.
You can use for example (xm)grace: http://plasma-gate.weizmann.ac.il/Grace/doc/UsersGuide.html#correlation/covariance which is free.
Depending on the MD code you have used, there may be a tool already included like 'g_analyze' which comes with GROMACS: http://manual.gromacs.org/online/g_analyze.html
Autocorrelation function is implemented in many (commercial/open) software.
You can use for example (xm)grace: http://plasma-gate.weizmann.ac.il/Grace/doc/UsersGuide.html#correlation/covariance which is free.
Depending on the MD code you have used, there may be a tool already included like 'g_analyze' which comes with GROMACS: http://manual.gromacs.org/online/g_analyze.html
Thank you for the kind reply Dr. Frank. By the way, I have about 250 lipids in a frame and have about 10,000 frames. So I need a code to process those data. note: I am using AMBER for my simulation.
You are welcome! Maybe I can help you..
Can you schematically describe the (file) format of your results?
With which kind of programming language you are familiar with?
Please be careful as angles are circular variables, you will run into trouble calculating linear correlation with them. You can transform them to linear space by doing a sin + cos transform effectively doubling your angular dimensions !!
I will recommend using the following package of time series analysis available at this link,
http://www.mpipks-dresden.mpg.de/~tisean/
Greetings to you Dr. Frank. Sorry for little late response. I shall explain briefly the format of the file that I have for analysis.
I have 64 lipids in one frame (single mono-layer) and in total I have 4000 frames. I calculated the tilt angle of every lipid from its bilayer normal for every frame and saved the angles in a file for each corresponding frame. So now I have 4000 files each contains 64 values representing each lipid in a mono-layer.
After that, I selected all angle values of first lipid and stored in new file- (example: angles-of-lipid1.dat). Then, all the angle for second lipid was collected and stored into another file (angles-of-lipid2.dat) and so on. In this way, I have total of 64 files with each file contain 4000 values from every frames for the corresponding lipid. Now I am trying to calculate the ACF.
I can do coding on FORTRAN 95.
Thank you.
Dear Abhinav Jain,
Thanks for your suggestion.
I will look into the link you provide.
Regards
Dear Abhinav Jain, could you suggest a way to transform the circular variable angle into linear variable? Any site or literature also will do. Thanks in advance.
How to calculate the reorientational dynamics of water around polymer chain in MD using GROMACS or by code?
- Badges
- Science topic