freebayes calls presumably homozygous variants as heterozygotes

Question

[johanna.pieplow]

Hi Everybody,

I ran freebayes to detect variants in multiple samples and now I saw that sometimes at a site there is only one reference observation and 10 or more alternate observations but still the sample is called heterozygous, even though it is more likely to be homozygous for the alternative allele in fact. Often also the reference observation has quite low quality in comparison to the alternative observation.

I used the --min-alternate-total and --min-alternate-fraction parameters to set a threshhold for when to consider an alternative allele, but is there a way to set a threshhold like this for the reference allele as well or any other way to reduce this behaviour?

[frankiefattorini]

Hi, I have also had this problem but haven't found a solution. Have you found anything since posting this?

[virginia.baraja]

Hello, I have faced the same problem related to the designation of heterozygous genotypes when the alternate allele frequency is too low (expected hom-reference genotype) or too high (expected hom-alternate genotype). Is there any solution to deal with that issue?