Allpath-LG assembled contig fasta file contain "B"

[intawat]

I us Allpath-LG to assembly a genome and I got "B" letter in fasta file as be low


TTAAACTTGAAATACCCTATCATCCAGGCCCCTATGGCGGGGGTCACGACTATTGAAATGGCCGCTAAGGCTTGTATTGC
GGGBGCCATAGCTTCACTACCCCTATCCCACTTAGACTTCAGAAAGGTCAATGATATTGAAAAGCTTAAACTGATGGTTT
CACAATTCAGAGATCAAGTAGCCGATGAATCTTTAGAGGGCAATCTCAACCTAAACTTTTTTTGCCATRATATCGTTGAT

What happen ?

[maubp]

Assuming Allpath-LG supports this, it could be an IUPAC ambiguity code meaning C or G or T.

Or, it could simply be an error - that kind of thing can happen with faulty RAM for example. Try running memtest on your machine?

[AdrianP]

I us Allpath-LG to assembly a genome and I got "B" letter in fasta file as be low


TTAAACTTGAAATACCCTATCATCCAGGCCCCTATGGCGGGGGTCACGACTATTGAAATGGCCGCTAAGGCTTGTATTGC
GGGBGCCATAGCTTCACTACCCCTATCCCACTTAGACTTCAGAAAGGTCAATGATATTGAAAAGCTTAAACTGATGGTTT
CACAATTCAGAGATCAAGTAGCCGATGAATCTTTAGAGGGCAATCTCAACCTAAACTTTTTTTGCCATRATATCGTTGAT

What happen ?

grep -c "B" <file with contigs>

Is the output more than 1?

[acgt]

I have encountered this too. I have 46 Bs in my assembly and several other IUPAC ambiguity code symbols (e.g. S, V, H), which appear to be more common in my Allpaths-LG assembly compared to other assemblers' output.

Does anyone know if there are options to change the base calling in Allpaths? Or at the very least, what algorithm the program is using to decide an ambiguous base? I think the program is being too conservative.