@Suji

In Gromacs 2023, it doesn't generate xpm file during gmx dssp. It only generates dat file of secondary structure.

Hello Ankur Chaudhuri you could simply take t he .dat file and run through Tcl script for plotting ss. In NAMD tutorial you can find Tcl script to calculate the same.

The code shall just take your .dat info as input for plotting.

Hope this helps.

Best

Shubhashree

Hello Shubhashree Barik. Thank you for your response.

Could you please provide me the tcl script and tell me the procedure?

I have faced the same report. .dat file can be opened by a text editor like notepad++. Import data into excel then you can count the occurrences of each character in each row to give a proportion of the protein in each secondary structure at each time point. You can plot these proportions over time to show how the secondary structure elements evolve. In this file S is for beta-sheet, ~ is for coil, and P is for turn.

Ankur Chaudhuri Here, I have uploaded a photo of my calculation so you can use it for you own. You simply need to count the character in each "row". Just open .dat file and copy-past it into an excel file. Use the formula in the photo, and only replace S, ~, or P for each type of your calculation. Note that each row is a frame.

Thank you very much. @Alireza Mohebbi

Hello Ankur Chaudhuri

I have developed a method that should be of help. This involves using Pandas to parse the .dat file and computing the frequency of each secondary structure type at each residue over the whole trajectory. This allows us to plot each secondary structure against the sequence, enabling visualization of both uniform and variable regions. You can see and example in the attached DSSP-plot.png. It's also possible to summarize this variability with a single score to highlight areas with highly variable secondary structure, as shown in windowed-variability.png.

I have posted an initial version of the code for this process on GitHub, available at: https://github.com/stuartmac/gmx-dssp-analysis. To get started, please refer to the prototype.ipynb Jupyter notebook in the 'notebooks' directory. This notebook should serve as a suitable guide for generating trajectory plots from your ss.dat files, and also demonstrates how I produced the attached plots.

You might note that I've simplified the 7-state DSSP to a 3-state model. This is a fairly common practice, but you can easily adjust the code to visualize all 7 states if you prefer.

On a side note, I noticed a couple of values ("~" and "P") in the data that I'm unfamiliar with. Do you happen to know what they represent?

I hope this solution helps, let me know if you have any questions.

@ Stuart Alexander MacGowan. Thank you so much for your reply. "~" and "P" represents coil and turn respectively. These automatically generated from the gmx dssp module of Gromacs.

"gmx do_dssp -f mdrun.xtc -s mdrun.tpr -o ss.xpm"

Using this command you can obtain .xpm file, which can be converted to .eps and viewed easily.

Adwaita Nair

Thank you very much.

This is not working in Gromacs 2023. In this version, "gmx dssp" generated ".dat" file.

Stuart Alexander MacGowan

Thank you so much for your reply. "~" and "P" represents coil and turn respectively. These automatically generated from the gmx dssp module of Gromacs 2023.

You're very welcome Ankur Chaudhuri ! Thanks for the information about the GROMACS DSSP types. I've updated my notebook on GitHub to reflect this (https://github.com/stuartmac/gmx-dssp-analysis/tree/v0.1.0).

Interestingly, the most variable region now looks to be towards the C-term, where there is an oscillation between helix and coil! What do you think?

Please feel free to fork the repository and adapt it to suit your specific requirements.

What does = in the dssp.dat file denote as a secondary structure type?