Beauti 2.0.2 not accepting .nex file about beast2 HOT 6 CLOSED

This may be due to the nexus file using X to represent missing values. Both nucleotide and amino acid data types should probably be able to deal with this.

from beast2.

X is not a valid code for missing data in DNA/RNA. We should stick to established codes. It might be valid for amino acid sequences but that should be confirmed before we allow it.

Perhaps a better error message (i.e. popup saying illegal character in sequence) would be sufficient?

Cheers
Alexei

Sent from my iPhone

On 17/02/2014, at 8:28 AM, rbouckaert [email protected] wrote:

This may be due to the nexus file using X to represent missing values. Both nucleotide and amino acid data types should probably be able to deal with this.

—
Reply to this email directly or view it on GitHub.

from beast2.

http://en.wikipedia.org/wiki/FASTA_format suggests X is used as "masked" for nucleotide, and "any" for amino acid coding. Since "X" is not used for anything else in nucleotide and amino acid encoding in BEAST, we can keep to the same convention as for FASTA, and interpret and "X" the same as a question-mark.

from beast2.

Lets not introduce our own conventions. IUPAC is the convention for nucleotides.

Alexei

Sent from my iPhone

On 24/03/2014, at 9:20 AM, rbouckaert [email protected] wrote:

http://en.wikipedia.org/wiki/FASTA_format suggests X is used as "masked" for nucleotide, and "any" for amino acid coding. Since "X" is not used for anything else in nucleotide and amino acid encoding in BEAST, we can keep to the same convention as for FASTA, and interpret and "X" the same as a question-mark.

—
Reply to this email directly or view it on GitHub.

from beast2.

How do you suggest to interpret the 'X' in nucleotide alignments allegedly produced by clustalX?
At least the FASTA parser needs converting 'X's.

from beast2.

@alexeid seems to be right regarding X not being used this way under the IUPAC nucelotide convention (in fact, it represents "xanthine"). It does represent ambiguity in amino acid sequences though.

Still, if X is defined in FASTA this way, shouldn't the parser read it this way?

from beast2.