8.9 years ago by
Predicting the effects of mutations on the structure and properties of a protein is a very difficult and inaccurate business. The most important thing to bear in mind is that your predictions are very unlikely to be meaningful in terms of "biological reality."
Molecular dynamics has been used to study point mutations; a simple web search for those terms brings up quite a lot of publications. For example:
Insight into a Novel p53 Single Point Mutation (G389E) by Molecular Dynamics Simulations
However, MD is not a trivial exercise. It requires a lot of computing power (normally, access to a cluster) and expertise (in setting up the correct parameters). Furthermore, simulations are generally limited to picosecond timescales which are likely to be too short to observe protein melting in most cases. If you do want to experiment with MD, GROMACS is good for beginners.
Some researchers have also looked at whether primary sequence can be used to predict melting temperature, e.g.:
There is some indication that sequence dipeptide composition is a reasonably good predictor. This is also the basis of the instability index, used by predictive tools such as ProtParam.