I have a question about the error in the homopolymer region.
I can understand there is a possibility for a deletion error like this (fig.1), but is there a possibility for an insertion error like this (fig. 2)?
It is possible. The culprit would likely be sequence error from the very inefficient Sequence-By-Synthesis method, which may have introduced a G base in some reads. The possibility for mis-alignment from another read [originating from elsewhere in the genome], which genuinely contains a G base, is also possible.
Thank you Kevin!