autonomously diagnoses rare diseases
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.
Moon has been beta tested and validated at:
The proof of the pudding is in the eating: we sent solved cases to test the software and in each case the causal mutation was found in no time.
Prof. dr. Geert Mortier, MD
Chairman, Department of Medical Genetics
Antwerp University Hospital (UZA)
Geneticists upload NGS data as a standard VCF file (e.g. directly from GATK) and they enter the patient's symptoms, gender and age.
Moon takes this input and, using proprietary artificial intelligence algorithms, outputs a diagnosis in under 3 minutes. The results can then be verified by a geneticist.
We tested Moon using 100 real-life single exome cases (without family data) that had previously been 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 directly solved 90% of these cases, with the causal variant appearing at the top of Moon’s list of candidate variants. In 100% of the cases, the causal variant was listed in Moon's top 10. For cases with family data available, Moon performs even better.
Moon is aimed at developers of variant analysis software. Moon can be embedded into variant analysis software to offer a more intelligent way of diagnosing rare diseases. In addition, Moon is available for country-wide genome programmes as well as high-throughput genome centres.
Want to learn more?contact us