Alternative tools replace MutDB to predict consequences of deleterious SNV

Question

[Shicheng Guo]

Any Alternative tools replace MutDB to predict consequences of deleterious SNV? It seems MutDB is already down.

Prediction of phenotypic consequence for deleterious coding nSNPs The phenotypic consequences of deleterious coding nSNPs were predicted using MutDB (http://www.mutdb.org/cgi-bin/mutdb.pl; access date: May 20, 2013), a tool that integrates publicly available databases of human genetic variation with molecular features and clinical phenotype data [20]. Gene symbol (‘TP53’) was used as a search term.

Conclusion: One Yes, then Yes. Means any method think it is missense mutation then it is or else you might lost your target.

This Link is quite useful:

https://brb.nci.nih.gov/seqtools/colexpanno.html#dbnsfp

Answer 1

[Collin]

You could use scores from the VEST algorithm (PMID: 23819870). It's pretty easy to score SNVs by just submitting them to the CRAVAT webserver (http://www.cravat.us/CRAVAT/, PMID: 29092935). In one of the latest papers to benchmark performance, VEST is among the best (PMID: 29179779).

[Shicheng Guo]

Not so good, my expected missense mutation is not significant with this method.

Do you think the idea of VEST is too simple? just use Random Forest and run a machine learning?

The Variant Effect Scoring Tool (VEST) is a new method for prioritizing missense mutations that alter protein activ- ity. VEST uses a supervised machine learning algorithm, Random Forest [22,23], to identify likely functional mis- sense mutations. The training set is a positive class of mis- sense variants from the Human Gene Mutation Database and a negative class of common missense variants detected in the Exome Sequencing Project (ESP) population.

[Collin]

Could you clarify whether you are interested in germline mutations or somatic mutations? The table you include has many cancer-specific methods, but the semantic "deleterious SNV" would suggest you are interested in germline mutations.