Which To Consider As Non-Sense Mutation From Annovar Output?

Question

[ivivek_ngs]

Dear All,

I have a question regarding prioritizing the variants after annotating them with ANNOVAR. I am interested in selecting only the non-sense mutations. So according to the classical definition, a nonsense mutation is a point mutation in a sequence of DNA that results in a premature stop codon, or a nonsense codon in the transcribed mRNA, and in a truncated, incomplete, and usually nonfunctional protein product (wiki). In annovar out, like the exome_summary.csv file when you filter out the exonic column you can see for non-sense mutations you have both stop-gain and stop loss listed. So for non-sense mutation extraction should I consider both or only the stop gain? Can anyone throw me some idea on this, it will be highly appreciated.

[ivivek_ngs]

In annovar website the stop loss definition is a nonsynonymous SNV, frameshift insertion/deletion, nonframeshift insertion/deletion or block substitution that lead to the immediate elimination of stop codon at the variant site. So in ideal situation this should not be considered as non-sense mutation as for non-sense mutation it is important to encounter a stop codon. So as per the definition I would only consider the stop gain as non-sense mutation. Does that make sense?

Answer 1

[Ashutosh Pandey]

I would consider both stop codon gains and losses as non-sense mutations. The reason being is in genomics, things are relative as variants are discovered by aligning reads against the reference genome. So a stop codon loss in the non-reference genome can also be interpreted as stop codon gain in reference genome. We don't know which transcript variant is the functional one or in other words, if the reference genome lost a stop codon or the non-reference genome gained a stop codon.

How to know which transcript variant resembles more to wildtype ?

You can perform multiple sequence alignment of the gene sequence against different species and check the consensus. For example, if you have a strain of mouse and you found a stop codon gain in some gene when compared to the reference mouse genome (C57BL/6J). Now you can take both the reference version of the gene and the other version with the stop codon gain and compare it against human, chimpanzee, rat etc. The copy that resembles to most of the species can be assumed to be the wildtype. This method can not be used to know which transcript variant produces functional protein.