Intended use

Product description

Moon is decision-support software for interpretation of next generation sequencing (NGS) data in the context of rare disease diagnostics. The cloud-based software is intended to be used by human genetics professionals only in a professional environment (eg. clinical genetics lab) and serves as an aid in identifying genetic variants of relevance for rare genetic diseases of Mendelian etiology. To this end, the device consists of both an Automated analysis and a Manual filtering pipeline.

In addition to providing decision-support for NGS interpretation, Moon stores the uploaded genetic data and patient information, provides a way to classify variants and store these classifications, provides the possibility to draft a report containing the variants of interest and provides a way to automate autonomous reanalysis of samples of choice.

Workflow

An overview of the recommended workflow in Moon is displayed in the diagram below:

Open

As input data, Moon requires NGS data, either from targeted panels, whole exome or whole genome data, in VCF format. In addition, Patient info is required (gender, age of onset of symptoms, clinical features of a patient, family relations).

Moon annotates the uploaded VCF file(s) with a wide variety of relevant Annotations (eg. allele frequency info, known pathogenicity, segregation, phenotype overlap etc.) needed to identify variants of interest.

Based on the combination of these annotations and the inputted patient/family information, Moon’s automated pipeline filters down the variants in the VCF file to a list of relevant variants. Finally, the variants that remain after filtering, are ranked based on the combination of variant annotations. The Automated analysis only considers variants in genes for which a gene-disease association has been published in scientific literature.

The output of the automated Moon pipeline, called the shortlist, comprises of a ranked lists of annotated variants, which were selected from the originally uploaded VCF(s). Critical assessment of these outputted variants by human genetics professionals is then required. The manual assessment of the results is facilitated by the annotations and links provided in the Moon UI. In case no conclusive diagnosis can be reached based on the results of the automated pipeline, manual filtering of the data must be conducted. Performance of the automated pipeline has been reported to be in line with manual state-of-the-art interpretation, with 99% precision and 97% recall for retrospective analyses of cases with known diagnoses.

Moon’s Manual filtering functionality displays a wide range of annotations of all rare variants in the VCF and provides the mean to manually filter the data according to custom filter settings (eg. zygosity, segregation, in silico gene panels, gene names, chromosomal regions, population frequency, variant quality…). The manual filtering module is the state-of-the-art method for NGS interpretation, and allows the user to either confirm the results from the automated pipeline, or to select variants of relevance for the patient that were not outputted by the automated pipeline (eg. variants outside of the scope of the automated pipeline such as variants in new disease genes; variants missed by the automated pipeline).