An AI-based approach driven by genotypes and phenotypes to uplift the diagnostic yield of genetic diseases.

Identifying disease-causing variants in Rare Disease patients' genome is a challenging problem. To accomplish this task, we describe a machine learning framework, that we called "Suggested Diagnosis", whose aim is to prioritize genetic variants in an exome/genome based on the probability of being disease-causing. To do so, our method leverages standard guidelines for germline variant interpretation as defined by the American College of Human Genomics (ACMG) and the Association for Molecular Pathology (AMP), inheritance information, phenotypic similarity, and variant quality. Starting from (1) the VCF file containing proband's variants, (2) the list of proband's phenotypes encoded in Human Phenotype Ontology terms, and optionally (3) the information about family members (if available), the "Suggested Diagnosis" ranks all the variants according to their machine learning prediction. This method significantly reduces the number of variants that need to be evaluated by geneticists by pinpointing causative variants in the very first positions of the prioritized list. Most importantly, our approach proved to be among the top performers within the CAGI6 Rare Genome Project Challenge, where it was able to rank the true causative variant among the first positions and, uniquely among all the challenge participants, increased the diagnostic yield of 12.5% by solving 2 undiagnosed cases.

Diagnostic application of a capture based NGS test for the concurrent detection of variants in sequence and copy number as well as LOH.

Whole exome sequencing (WES) has made the identification of causative SNVs/InDels associated with rare Mendelian conditions increasingly accessible. Incorporation of softwares allowing CNVs detection into the WES bioinformatics pipelines may increase the diagnostic yield. However, no standard protocols for this analysis are so far available and CNVs in non-coding regions are totally missed by WES, in spite of their possible role in the regulation of the flanking genes expression. So, in a number of cases the diagnostic workflow contemplates an initial investigation by genomic arrays followed, in the negative cases, by WES. The opposite workflow may also be applied, according to the familial segregation of the disease. We show preliminary results for a diagnostic application of a single next generation sequencing panel permitting the concurrent detection of LOH and variations in sequences and copy number. This approach allowed us to highlight compound heterozygosity for a CNV and a sequence variant in a number of cases, the duplication of a non-coding region responsible for sex reversal, and a whole-chromosome isodisomy causing reduction to homozygosity for a WFS1 variant. Moreover, the panel enabled us to detect deletions, duplications, and amplifications with sensitivity comparable to that of the most widely used array-CGH platforms.