Analyzing REMD trajectories to understand effect of temperature on protein structure

User discussions
1

GROMACS version: 2020.5
GROMACS modification: No
Hi,

I am currently working on analyzing temperature replica exchange molecular dynamics (T-REMD) simulations for a protein system, and I would greatly appreciate your guidance and suggestions on the analysis.

In my simulations, I have data for 44 replicas spanning a temperature range of 323–363 K. The available files include:

  • RMSD trajectories (rmsd_0.xvg to rmsd_43.xvg)

  • Radius of gyration trajectories (rg_0.xvg to rg_43.xvg)

  • DSSP secondary structure data (dssp_0.xpm to dssp_43.xpm)

  • Corresponding temperature information (temps.txt)

My objective is to gain insights into the thermostability of the protein and to determine up to what temperature the protein structure remains intact. I have tried various plotting methods to get insights into it but I am not able to get the desired resoults.

I would be very grateful if you could advise me on:

  1. The most suitable ways to analyze these datasets to assess structural stability across temperatures.

  2. How best to use the RMSD, Rg, and DSSP results to quantify the point at which the protein starts to unfold.

  3. Any additional analyses you would recommend to characterize the temperature-dependent structural transitions.

Your expertise in this area would be invaluable in helping me interpret the results correctly and extract meaningful insights about the protein’s thermostability.

Thank you very much for your time and consideration. I look forward to your advice guys.

2

Hi @ashish_bioinfo ,

in my opinion, T-REMD is not the right approach to test the thermostability of a protein. In T-REMD, the individual simulations are not independent from each other, as structures are regularly exchanged between replicas at different temperatures (depending on the temperature spacing). This makes it difficult to determine the specific temperature at which unfolding occurs, if it happens at all because the structure may only face for a short time high temperatures. Instead, T-REMD is more useful for observing structural rearrangements in the protein that may be difficult to capture in a long unbiased simulation at room temperature.

For your specific problem, I would recommend running a set of annealing simulations, where the temperature is slowly ramped from 323 K to 363 K using the GROMACS annealing protocol. That said, other users may have different perspectives, since this is not my main area of expertise.

Best,

Marius