After some preliminary analysis, we find one particular variant (179248034 on chromosome 5) in gene SQSTM1 that is likely to be associated with a phenotype. Let us try to find more information about this variant, which is in a variant table named MyVar in our project.

Gene and mRNA

We first would like to download and use the ref gene database to our project. To learn what fields are available in this annotation database, we can use command vtools show annotation to print the details of it.

vtools use refGene
vtools show annotation refGene

The following command output the corresponding records in refGene database for this variant:

vtools output myVar chr pos ref alt txStart txEnd cdsStart cdsEnd exonCount name2
5       179248034       C       T       NM_001142298    179233388       179265077       179250005       179263593       9       SQSTM1
5       179248034       C       T       NM_001142299    179234003       179265077       179250005       179263593       9       SQSTM1
5       179248034       C       T       NM_003900       179247842       179265077       179247937       179263593       8       SQSTM1

If we use the ccdsGene database, we would get:

vtools use ccdsGene
vtools output myVar chr pos ref alt ccdsGene.txStart ccdsGene.txEnd ccdsGene.cdsStart ccdsGene.cdsEnd ccdsGene.exonCount ccdsGene.name2
5       179248034       C       T       CCDS34317.1     179247937       179263593       179247937       179263593       8

We had to use full name ccdsGene.txStart etc to avoid ambiguity because these fields also exist in the refGene database.

The results show that

  • This variant locates in consensus CDS gene CCDS34317.1, with common name SQSTM1.
This gene encodes a multifunctional protein that binds ubiquitin and regulates activation of the nuclear factor kappa-B (NF-kB) signaling pathway. The protein functions as a scaffolding/adaptor protein in concert with TNF receptor-associated factor 6 to mediate activation of NF-kB in response to upstream signals. Alternatively spliced transcript variants encoding either the same or different isoforms have been identified for this gene. Mutations in this gene result in sporadic and familial Paget disease of bone.
  • This gene can transcribed two three mRNAs NM_001142298, NM_00142299 and NM_003900, which corresponds to CCDS id CCDS4735.1 (first two), and CCDS34317.1. Searching in CCDS gene only displays CCDS34317.1 because this variant falls out of the coding region of CCDS4735.1.

Which exon?

Now we know that this variant is within the coding region of CCDS34317.1, is it in one of the exon region? We can use the ccdsGene_exon database to find this out.

vtools use ccdsGene_exon
vtools output MyVar chr pos ref alt ccdsGene_exon.exon_start ccdsGene_exon.exon_end                        
5       179248034       C       T       179247937       179248141

So we know this variant stays in an exon starting from 179247937, according to here, this is the first exon of this gene.

Potential of damaging?

Because this variant belongs to one of the CCDS genes, we can find the SIFT score and PolyPhen2 (and other scores) of this variant from the dbNSFP database.

vtools use dbNSFP
vtools output MyVar chr pos ref alt SIFT_score polyphen2_score
5       179248034       C       T       0.49    0.893

This database also tells you how this mutation changes the aminoacid:

vtools output MyVar chr pos ref alt aaref aaalt refcodon codonpos aapos
5       179248034       C       T       A       V       GCG     2       33

So this variant mutates C in GCG to T, and changes aminoacid A (GCG) to V (GTG).

Now, let us get the aminoacid sequence from the CCDS gene page, and send it to the PolyPhen 2 website, we need to fill in

  • Protein identifier: leave blank or use name NP_003891.1
  • Protein sequence: copy and paste with leading line >NP_003891.1 or leave blank
  • Position: 33
  • From A -> V

Note that we can use either protein name or sequence, but not both.

The results show that

RecName: Full=Sequestosome-1; AltName: Full=EBI3-associated protein of 60 kDa; Short=EBIAP; Short=p60; AltName: Full=Phosphotyrosine-independent ligand for the Lck SH2 domain of 62 kDa; AltName: Full=Ubiquitin-binding protein p62; LENGTH: 440 AA

HumDiv: (preferred model of evaluating rare alleles, dense mapping and analysis of natural selection)
This mutation is predicted to be POSSIBLY DAMAGING with a score of 0.915 (sensitivity: 0.81; specificity: 0.94)

HumVar: (preferred model for diagnostics of Mendelian disease)
This mutation is predicted to be BENIGN with a score of 0.399 (sensitivity: 0.84; specificity: 0.78)

The PolyPhen2 score obtained from PolyPhen2 website differs from what is reported by dbNSFP, which used a previous version of PolyPhen2 to generate scores for these variants.