Hi,
Is there any good prediction tools that can predict deliriousness of nonsense variants?
Im trying to prioritize germline variants that might be potentially pathogenic to the disease that I'm investigating. After filtering for rare variants, and my genes of interest, i still have quite a number of nonsense variants.
When I check some of the nonsense variants, the stop codon is gained not too far away from the canonical stop codon and thus it "might" be less deleterious as compared to a stop codon that is gained near a start of a transcript?
I came across dbNSFP database and I could see some tools could predict deleteriousness of nonsense variants such as MutationTaster, LRT, FATHMM etc. How reliable are these prediction tools ?
Some of nonsense variants are indeed non-pathogenic but how can I predict deleteriousnesss for unknown nonsense variants?
I would say, in many cases this is search in some particular database:
https://www.hindawi.com/journals/tswj/2013/675851/
Prediction of Deleterious Nonsynonymous Single-Nucleotide Polymorphism for Human Diseases
“Many popular databases present useful information of nsSNPs. Particularly, as shown in Table 1, deleterious nsSNPs are mainly collected in four databases: the Online Mendelian Inheritance in Man (OMIM) database [22], the Human Gene Mutation Database (HGMD) [12], the UniProt/Swiss-Prot database [13], and the Human Genome Variation database (HGVbase) [14]. Other popular databases like the single-nucleotide polymorphism database (dbSNP) [15], the Protein Mutant Database (PMD) [16], and the database for nonsynonymous SNP’s function prediction (dbNSFP) [9] are also important for collecting nsSNP data (also shown in Table 1).”
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC3516590/
Deleterious- and Disease-Allele Prevalence in Healthy Individuals: Insights from Current Predictions, Mutation Databases, and Population-Scale Resequencing
“Inheritance in Man and Human Gene Mutation database [HGMD])1 and low-penetrance variants contributing to complex disorders (National Human Genome Research Institute [NHGRI] Catalog of Published Genome-wide Association Studies) attest to the progress made to date.”
http://www.ensembl.org/info/genome/variation/predicted_data.html Ensembl Variation - Predicted data
Thank you for the information,
However, most of the tools are designed to predict nonsynonymous missese based on multiple sequence alignment i.e variants at conserved site are likely to have larger functional impact. And also some other algorithms that make use protein biochemical properties, domains etc.
None of the tools actually addressees how a nonsense variant could be more deleterious than the other nonsense variants. I could predict deleteriousness of nonsense variants if they are presents in databases such as OMIM, HGMD etc, but how about nonsense variants that are not in those databases?