Questions about running MEGA phylogenetic analysis with mitrochondrial transcription factor TCF7L2

hey everyone,

I was bored today so for shits and giggles i compiled the same TCF7L2 sequence for 4 animals off of the NIH ( African Elephant, Emu, Saltwater croc and anole lizard) to compile into one FASTA file and then run through a phylogenetic analysis. Expecting to find the archosaurs to group together, i actually found that the elephant and emu grouped instead, with crocodiles sometimes being the outgroup to the others or sometimes forming a sister clade to the anoles. Theres a very good chance i just fucked up the process but i tried several different approaches ( Maximum Likelihood/Parsimony etc) but assuming i didnt  I was wondering why this is and did some digging, but only got some unsatisfying results so far. Apparently TCF7L2 is associated with a signaling pathway known as "Wnt" which can control anything from glucose levels (which may be tied to metabolic rate, so grouping of endotherms (?)) and heat based sexual determination ( as seen in crocs) but i kinda hit a wall. Anyone know any good sources or explanations for this or could point me in the right direction? thanks !