Hi,
I'm studying molecular evolution of a protein complex in prokaryotes, I have a tremendous fasta file that contains thousands of aa sequences of this protein from NCBI ref-seq protein database.Now I'd like to do the multiple alignment first and analyse the phylogeny.
I used MUSCLE and FastTree before, but the "tree" created doesn't make sense to me, the evolutionary pattern is not clear, out of curiosity, I'd like to see if the phylogenetic tree is accurate or not, so I picked two nodes that lie close to each other on the tree to do the pair-alignment in BLAST, but it turned out that they just share 35% identity and the coverage is very low.
I'm wondering is there any good methods to have a better phylogeny analysis? Am I on the right track? or just some technical problems, like did I miss any step? I haven't done the gblock and any trimming of the aa sequence, because I thought that it could give me a "better but less true" results compared with what they really are. But not sure if I understand correctly.
