Knowledge base

The Knowledge Base (KB) allows to create and maintain a database of variants with known clinical classifications for your lab account. These classifications are annotated by Moon in future analyses, and these annotations are used to improve automated filtering and ranking of the results. In addition, these KB classifications can be used as filter setting in the Filter view.

Any information in the KB is only used within your Moon lab account and not shared with other Moon users outside of your lab account.

The KB can be accessed through the Lab menu at the top left of the screen.

The overview page lists all SNVs currently included, with the genome build in which these were entered, and allows you to search within the KB based on variant position or gene name.

Clicking a particular entry shows more details of that particular variant including:

  • Classification: Benign - Likely benign - Variant of Unknown Significance (VUS) - Likely Pathogenic - Pathogenic

  • Note: When editing a KB SNP, a note can be added as free text.

  • Source: When adding or editing a KB SNP (either manually or through the API), a Source indicating the origin of the SNP KB entry (e.g. HGMD) can be added as free text. ‘Manual curation’ is set as source by default if manually adding or updating a KB classification.

  • References to literature: These can be added by simply pasting the PubMed ID or PubMed URL of the paper of interest, and then clicking 'Add publication'.

  • History: All changes in the classification of the variant, and information on when it was updated and who changed it, are listed in this section.

  • Samples: All samples in which the specific variant was shortlisted are listed. Clicking the sample names brings you to the most recent analysis results in which the variant is shortlisted.

Adding variants to the KB

SNVs can be added to your lab's KB in several ways:

  • KB: On the KB overview page, you have the possibility to manually add variants. The following info is required to create a KB entry: genomic location, genome build indication, reference allele, alternate allele and classification. The gene name and a note can be added optionally.

  • Analysis (Shortlist or Filter View): You can save a variant to your lab's KB from within a sample, by clicking the KB icon at the right of the variant overview bar. When clicking the KB icon, you will be redirected to the KB, where additional info can be entered. When adding a variant from within a sample, the genome build is automatically assigned based on the genome build assigned for the sample the variant was detected in. After adding a Shortlisted or Filtered variant to the KB, the KB icon remains visible with an indication that the variant has been added to the KB, and links out to the corresponding KB entry.

  • API: If you would like to import an external KB to your lab's account, you can use the Moon API.

Automated filtering based on KB classification

Moon will take into account clinical classifications of your KB SNVs for generation of its results through the automated analysis pipeline as follows:

  • Benign/Likely benign: Variants with these classifications will be filtered out of the automated pipeline. These variants can of course still be reviewed in the Filter view.

  • Pathogenic/Likely pathogenic: Variants with these classifications will receive a higher ranking in the shortlist. Phenotype-overlap with the patient remains required in order for the variant to be considered for the shortlist. The clinical classification as indicated in the KB, will be displayed in the overview table and variant card in the results.

  • Variants of unknown clinical significance (VUS): A VUS classification in the KB has no effect on filtering or ranking of variants. The VUS classification will be annotated as such in the overview table and variant card.

After adding a variant to the KB or updating its classification, the sample needs to be reanalysed for the annotation to be displayed. In the Shortlist, a message is shown if there were updates to the KB classifications of currently shortlisted variants, prompting the user to reanalyse the sample.

KB classifications are only annotated for samples of which the genome build is in accordance with the genome build of the variant in the KB.