Loudmouth: such fast sequencing could make molecular phylogeny much more useful, because having lots of gene sequence on hand tends to improve the resolution and reliability of molecular-phylogeny methods. This could help resolve a lot of riddles of evolution where molecular methods still do not give clear results, like early protist evolution.
PE 1: It may indeed be possible to estimate the number of change-of-function mutations from comparisons of genes. Those with changed functions will likely have more differences than one would predict from genes known to have unchanged functions.
PE 2: That may certainly be possible -- in fact, that's been done for some proteins, like a vision pigment of the shared ancestor of birds and crocodilians.
PE 3: One would have to know how to translate genes into shapes for that, but in principle, it ought to be possible.
As to the universal common ancestor, there has indeed been some work on that. Here are two references:
Brooks DJ, Fresco JR, Lesk AM, Singh M.
Evolution of amino acid frequencies in proteins over deep time: inferred order of introduction of amino acids into the genetic code.
Mol Biol Evol 2002 Oct;19(10):1645-55
Here in
PubMed
(Finds that the common ancestor's proteins were enriched in the amino acids that are easier to make prebiotically)
Davis BK.
Molecular evolution before the origin of species.
Prog Biophys Mol Biol 2002 May-Jul;79(1-3):77-133
Here in
PubMed
(tracks some proteins back before the common ancestor)
Ferredoxin, an iron-sulfur protein involved in electron transfer (an important metabolic process) is the oldest one; it has a negatively-charged "tail" that can attach to a mineral surface. This means that ferredoxin can be older than well-defined cells, suggesting a Haeckelian
Urschleim phase.
Finally,
LUA -- last universal ancestor
LCA -- last common ancestor
LUCA -- last universal common ancestor