Infering phylogenetic tree from MSA with MEGA ?

Question

[Picasa]

Hello,

I have aligned a set of sequence with MAFFT. Unfortunately the alignment is not perfect (gap) and

1) I am wondering if I need to filter somehing. Based on that paper :

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4538881/

It's apparently not good to filter. Did you have some experience about that ?

2) I would like to use MEGA 7 (or something else you can recommend) to compute the model that I sould use for maximul likelihood inference.

How can I import MAFFT result to MEGA ? The only thing I see is importing fasta file and align with muscle or clustal

Answer 1

[Tonor]

You should be able to load in you existing alignment and then save it in MEGA format for the tree. Align -> Edit/Build Alignment -> Retrieve Sequences from a file -> Select your .fasta alignment file. Then Data -> Export Alignment -> MEGA format -> Enter new output filename. Then in Phylogeny -> Construct/Test Maximum Likelihood Tree -> Select your MEGA format alignment.

An alternative to use after your MAFFT alignment is RAxML: http://sco.h-its.org/exelixis/software.html

[Picasa]

Good thanks.

Concerning my first question, do you any advice and/or resources (tutorial) about post processing a multiple alignment ?

[Tonor]

You can try trimAl: http://trimal.cgenomics.org

trimAl is a tool for the automated removal of spurious sequences or poorly aligned regions from a multiple sequence alignment