Introduction of a frameshift through insertion or deletion normally leads to truncation of the protein - the closer to the N-term of the protein is, the higher the probability that no functional protein at all is produced. Therefore, the effects of frameshift mutations are catastrophic unless the mutation is close to the C-terminus or followed by a compensatory frameshift mutation within a few codons. Since protein stability predictions are based on the assumption that the mutations does not change the overall structure of the protein, they cannot be applied to insertion/deletion mutations. mutations to a stop codon are yet a different story: the premature termination of protein synthesis may yield a nonfunctional protein, but depending of the producing organism, a low level of read-through may occur (suppression), leading to a small fraction of the protein being produced with a single point mutation rather than premature termination.
This protein have an active form as a proliferative mitogen. With the ProtParam the stabilty changes are notable. But i need other program more exclusive to examine other termodinamic values to predict changes in the protein stability.
Though @Annemarie Honegger has already explained in detail. But if you still insist, you can try this EnCOM web server that deals to predict thermodynamic stability of proteins and their mutations: