Is it possible to infer full ancestral protein sequences using MEGA? The reason I ask is that it seems like the only input to their ancestral reconstruction program is a newick tree, and not an MSA. Without knowing the actual sequences, I don't understand how it is possible to do any sort of reconstruction. If anybody has experience with this I would appreciate your feedback.
Yes, you're right, as with only a NEWICK tree, you can't predict ancestral states. Honestly, I hate using MEGA and would avoid using the program at all costs. I don't understand why anyone uses it, but, that's my opinion.
You can infer ancestral states using MESQUITE.