Infer ancestral sequence from a Multiple Sequence Alignment

Question

[anuragm]

I am trying to infer the ancestral state from a MSA that I have using MEGA. However, I am not sure how I will get a sequence as the ancestral state when there is no input in form of a MSA into the ancestral state reconstruction feature in MEGA. I am also not very sure how to interpret the output that comes in form of an excel sheet. I could not find a MEGA tutorial online for this. Could someone please point me towards online resources that could help me ?

[Michael]

See here for options: Is FastML a good server for reconstructing ancestral protein sequence?

[anuragm]

FastML crashed multiple times, just like it is mentioned!

[JulianZ]

Although this was posted a while ago, I thought I might offer a suggestion for anyone who stumbles across this in future - As far as I am aware, FastML can only handle less than 300 sequences (maybe even less than 200). So if you have more than this amount, try reducing the number.

Answer 1

[Biojl]

The easiest way to do it is to use the PRANK multiple alignment. Is as easy as activating the -showanc option when calling the command to build the MSA.

Notice that PRANK requires a phylogenetic tree as input, although it can also calculate it automatically using mafft (read the manual).