I was trying to draw whole genome phylogeny about 200 similar strains of bacterial genome. I align all the genome using Mauve aligner and got tree file and I can see it with any tree viewer. But when I gone through few recent paper I saw after alignment they usually extract the conserved block and draw the tree separately. Can anyone suggest me which will be optimal solution.
You can draw trees using several biological data. Usually people use highly conserved proteins like ribosomal proteins, align them, select blocks of alignments and draw a tree using this data. Another option is rRNA or other markers like insertion sequences and more. You should take into account that evolution of proteins might be different from evolution of rRNA or non-coding DNA.
That being said, every tree you build is valid, it just represents the data you used to build it. Since you are dealing with closely related bacteria it might not be such a bad idea to use DNA rather than highly conserved proteins.
Another issue is the algorithm used by Mauve, I'm not familiar with it so I can't say anything about it. You should have a way to validate your tree so you could trust it.
I believe Gblocks and trimal are the most widely used alignment trimmers. What software did they use in the few recent papers you saw? Since you're aligning complete genomes, the alignment is probably extremely long and thus not suitable for Bayesian tree-building approaches. RAxML would probably suit your needs well but if you want something much faster, then start with FastTree. Also, if you want a rooted tree, remember that you'll need an outgroup.
Thanks for reply. I saw the paper used Gblock to find conserved block and tree build by Figtree. If I got properly: according to your opinion my approach should be 1. align all genome using mauve 2. and draw tree using FastTree or RAxML. No need to find the conserved block after alignment.
You read the article wrong because FigTree is for illustration of trees that were made with whatever algorithm. You also read my post wrong. RAxML and FastTree can make tree files (e.g. nexus format), but you can't illustrate said files with these programs.
Thanks for reply. I saw the paper used Gblock to find conserved block and tree build by Figtree. If I got properly: according to your opinion my approach should be 1. align all genome using mauve 2. and draw tree using FastTree or RAxML. No need to find the conserved block after alignment.
You read the article wrong because FigTree is for illustration of trees that were made with whatever algorithm. You also read my post wrong. RAxML and FastTree can make tree files (e.g. nexus format), but you can't illustrate said files with these programs.
Sorry little modified tree was build by Mauve alignment and tree view by FigTree.
for fast tree I saw in manual they mention I have to use like
Could you please let me know in alignment file can I use a .aln file generated by mauve or else I have to build with other tool such as Mugsy?