How one can computationally analyse functional impact of frameshift mutation?
Note: You are only allowed to use freely available tools and databases.
How about Ensembl's Variant Effect Predictor?
There are a few instances of a SNP in a stop codon that extends the reading frame. There are also many examples of triplet repeats that alter the length of the gene, its mRNA and the translated protein product. These cases may best be analyzed with tools that predict protein secondary and/or tertiary structure.
I would also search the frameshifted sequence against an EST dataset (e.g. with BLAST) to see if the altered sequence is found, perhaps preferentially, in a cancer cell line over normal cells/tissue.
Lastly, there is some work coming from yeast (I don't have a reference on hand) that shows a peak in strength of the phenotype effect(s) of frameshift mutations for those placed in the middle of the protein. The rationale here being that very prematurely short peptides are functionally dead and quickly removed, while the nearly full-length proteins are likely to be tolerated and rather functional. The mid-length "half" peptides are much more likely to be deleterious.
Frameshift mutations often result in disruptions by giving rise to a premature stop codon or they will completely alter the AA sequence, thus they are easy to deal with: translate both sequences, you will see the effect on the AA level. Then one can search the disrupted AA sequence against PFAM, you can then try to find which portion of the protein domains remain intact in comparison to the protein family, and which ones get disrupted.
To bolster Larry's answer, in the case of pre-mature STOP codons, three things are likely to happen: 1. Non-mediated decay (NMD) 2. Non-sense protein 3. Selenoprotein formed.
As alluded to by Larry, NMD is most likely to occur in most cases resulting in mRNA destruction, thus no protein is produced. Returning to the question of effect: If the sequence is not known, you may run 6 - frame translation on the DNA sequence, obtain possible peptide sequences and try mutationtaser to ascertain if deleterious or not. Otherwise, assume deleterious effect in most cases whether or not the gene is translated.
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
May I ask why we are only allowed to use online tools and databases? If you have a problem that you need solved, you should use whichever tool is most appropriate, be it online or not.
In general it would be useful to know about commercial tools as well, at least if there are no free tools around. After all tool development also costs time (and in most cases that means money. Although there are often good reasons to use open source and thus extensible tools that you can precisely valuate. But if nothing else the commercial tools can at least show you what could be done.
By online tools I meant freely available tools and I completely agree if we have a problem it should be addressed it doesn't matter whether online or offline........