analyse rare disease NGS data
Moon is chosen by:
Moon is an AI-driven decision-support software for interpretation of NGS data in the context of rare disease diagnostics. Moon reduces the time needed for filtering and prioritising variants from days or weeks to mere minutes.
Diagnosing rare diseases can be like searching for a
needle in a haystack.
When interpreting exome data for rare disease diagnostics, geneticists must identify the one or two mutations responsible for the patient’s condition hidden amongst 40,000 variants. This process takes between 20 and 40 hours when performed by highly skilled staff using specialised software (Wenger et al. 2016). This is prohibitively slow for many clinical applications.
Moon changes all this.
Geneticists upload NGS data (SNV/CNV) as a standard VCF file (e.g. directly from GATK) and
they enter the patient's symptoms, gender and age of onset.
Moon takes this input and, using proprietary A.I. algorithms and a proprietary disorder model, suggests the causal variant in 2 minutes for WES and in 5 minutes for WGS. For each suggested candidate variant, a wide range of annotations are shown, thereby providing scientific evidence for Moon's choices. The results can then easily be verified and reported.
“Moon has accelerated our diagnostic process. It is not only fast but also very up to date: we diagnosed cases with pathogenic mutations in genes that were just published as disease genes. With Moon, yesterday’s publication is today’s diagnosis.”
Prof. dr. Frank Baas, MD
Head, Laboratory Diagnostic Genome Analysis
Leiden University Medical Center, The Netherlands
“Moon is a real added value to our diagnostic pipeline for NGS: very user-friendly, performant and up-to-date with the newest data from the recent literature.”
Prof. dr. Geert Mortier, MD
Chairman, Department of Medical Genetics
Antwerp University Hospital, Belgium
“In whole exome and genome sequencing, variant interpretation is a big challenge. Moon can provide the disease-underlying variant(s) with a few clicks in a couple of minutes, even among a large number of whole-genome variants. As a medical geneticist, I really appreciate Moon’s speed and easy HPO-based workflow. Moon helps us to diagnose rare diseases, presenting clinically important variants we should not miss.”
Dr. Gabor Matyas, PhD, FAMH Medical Genetics
Head, Swiss Foundation for People with Rare Diseases
Zentrum für Kardiovaskuläre Genetik und Gendiagnostik, Switzerland
At Invitae, we tested Moon in a pilot study of 150 exome cases previously solved by our expert geneticists. The cases were previously unknown to Moon, and the software had not been trained on the cases in any way. Moon alone correctly identified more than 97% of causative variants in less than two minutes per exome. Additionally, Moon has also demonstrated significant efficiency improvements at Invitae for its exome analysis: the number of incidental findings has been decreased by nearly half, without affecting the diagnostic yield, and the number of variants selected as candidates and for manual review has also decreased, reducing the time required to analyze a case by more than half.
Additionally, the performance of Moon's phenotype-driven pipeline continues to be established by a growing number of worldwide institutes, and has been reported in several peer-reviewed studies. Rady Children's Hospital validated Moon in 100 WGS samples as part of their rapid WGS workflow (Clark et al., 2019). The automated pipeline with Moon reached 99% precision on retrospective cases, and 100% precision on the prospective cases reviewed. In another study, Knight Diagnostic Laboratories (OHSU) validated Moon and used it for comprehensive AI-assisted exome reanalysis (O'Brien at al., 2022). Moon identified the causal variant in 97% (28/29) of known positive cases, and found the causal variant in 7 cases that were negative after initial analysis, contributing to a 9% increase in diagnostic yield.
See how Moon compares to Exomiser:View White Paper
Rapid exome sequencing and interpretation can be crucial in the acute care of infants with genetic diseases in neonatal and paediatric intensive care units. Moon can go from standard VCF to a shortlist of provisional diagnoses in a ground-breaking 2 minutes for WES data, and in 5 minutes for WGS data. As such, Moon can be instrumental for this new wave of time-sensitive NGS-based diagnostics. Moon’s fast analysis times contributed to a new world record for fastest genome diagnosis, set by researchers at Rady Children’s Hospital in San Diego.
By reducing manual analysis time from between 20 and 40 hours per exome interpretation (Wenger et al. 2016) to less than an hour, Moon can potentially save between 19 and 39 hours of labor. The additional expert time that becomes available by using Moon, can be highly valuable for in-depth analysis of difficult, unsolved cases.
Current software packages allow their users to define pipelines which are then applied to every exome analysed. However, because new gene-phenotype correlations are published every day, these pipelines quickly become outdated. Despite this, the gene panels used by these pipelines often remain untouched for months or even years. Moon is different in that it automatically scans the literature on a daily basis, integrating new scientific insights as they are published. As a result, Moon always gives you the best chance of reaching a diagnosis.
Since Moon can re-run analyses unsupervised, it's the perfect tool to re-run older analyses. Patients for whom no diagnosis is available today, might be diagnosable next month. Or next year. With Moon, infinite interpretation becomes a reality as re-running analyses is easy, fast and cheap.
While array-based methods have long been the standard to detect pathogenic CNVs, the relatively
low resolution only allows to detect CNVs of approximately 20,000 base pairs in size or larger. Since the
majority of CNVs in the human genome are smaller,
this means that about 80% of all human CNVs are missed with the traditional approach. NGS-based CNV detection
provides a solution to this problem, as CNVs as small as 100 base pairs can be detected with this method. Fast
interpretation of these called CNVs can now be performed with Moon.
Moon’s automated filtering and ranking algorithms quickly guide you to the relevant CNVs for your patient’s phenotype. Moon can even detect relevant combinations of SNVs and CNVs, as SNV and CNV analyses are performed simultaneously. Finally, the rich CNV annotations and automated reporting allow for easy manual review of the Moon results.
Moon is deployed in ISO27001 or HIPAA compliant data centres that are always located in the customer’s country. That way, genetic data remains in the country where it has been generated. Moon runs on dedicated physical servers that adhere to the highest security standards and are operated in compliance with HIPAA and GDPR. At the application level, Moon offers two factor authentication, making sure that only your lab members can access Moon.
Give Moon a try with your own samples.
Click "Try now" for a web-based demo and a Moon trial account that will allow you
to analyse your own WES or WGS data.