consensus sequence from BAM file

Question

[Ric]

Hi, I used samtools mpileup -uf chr01.fasta chr01.bam | bcftools call -c | vcfutils.pl vcf2fq > chr01-cns.fastq to make a consensus sequence from BAM file. However, it has used IUPAC characters. How would it be possible to choose a base with the most appearance at locus position rather than IUPAC?

Thank you in advance.

[WouterDeCoster]

What about heterozygous variants?

[Ric]

I do not care about them.

[WouterDeCoster]

So what do you do if a position has 10 reads A and 10 reads C?

[Ric]

Good question. What would error correction for PacBio do in this case?

[Gabriel R.]

I think WoulterDeCoster's question was about, what do you do if you have a heterozygous site with a 50/50 distribution.

[Ric]

I do not know what to do with 50/50 distribution. What would error correction for PacBio do in this case?

[Gabriel R.]

Well I hope that any error correction software does not "correct" true het sites :-) Why can't you use a small script that picks an allele at random at UIPAC ambiguous codes?

[Ric]

Hi, I ended up to write a script.

[Gabriel R.]

it this mt or nuclear?

[Ric]

it is nuclear data..