Hi all,


I´m having an issue with the latest version of Hyphy (0.99):
when reading amino acid data from a multi-fasta file the program recognizes it as if it were a nucleotide one.

But the same file converted to nexus by MEGA is correctly interpreted as aminoacid data.

The file I am having trouble is in:

Multimedia File Viewing and Clickable Links are available for Registered Members only!!  You need to Login

Do you have any guess about what might be causing this problem?

Best regards,

Dear Filipe,

Since FASTA does not provide a means to specify what type of data are present in the sequence file (NEXUS does, via the DATATYPE construct), HyPhy tries to make an educated guess about what the file contains. All files are assumed to be nucleotide by default, but if there are too many non ACGT characters (more than 50%), HyPhy assumes that the data are amino-acid sequences. This heuristic breaks down for very gappy alignments (which yours appears to be).

You can include the following HyPhy construct at the beginning of a FASTA/PHYLIP file to force the program to use the standard amino-acid alphabet:

[code]
$BASESET:BASE20

>seq1
...
>seq2
...
[/code]

Alternatively, using the NEXUS format avoids the issue altogether.

HTH,
Sergei