Table 2.
Combination of criteria for variant interpretation according to ACMG-AMP guidelines and modifications implemented in the internal ACMG-based classifier.
| BENIGN CRITERIA | ||
|---|---|---|
| ACMG CRITERIA | ASSIGNED BY VARSOME | EXCEPTIONS AND INDICATIONS |
| BA1 Allele frequency > 5% in Exome Sequencing Project, 1000 Genomes Project or Exome Aggregation Consortium | YES | Variants recommended in Ghosh et al. 2018 - Hypomorphic variants: NM_000350 (ABCA4):c.5603A>T (p.Asn1868Ile) NM_000372 (TYR):c.1205G>A (p.Arg402Gln) |
| BS1 Allele frequency greater than expected for disorder | YES (not evaluated if BA1 or PM2 are activated) |
Use STRONG as default. For AD diseases with high penetrance, the criterion can be used as STAND-ALONE evidence (sufficient to classify a variant as likely benign). |
| BS2 Observed in a healthy adult for a recessive (homozygous), dominant (heterozygous), or X-linked (hemizygous) disorder, with full penetrance expected at an early age | YES (not evaluated if BA1 or PM2 are activated) | Since VarSome retrieves information from the CGD database, in cases of known inheritance discrepancies, use the following gnomAD cutoffs * to include other variants in the selection: - AR/XL model: <3 homozygotes/hemizygotes in gnomAD exomes&genomes - AD model: <5 heterozygotes in gnomAD exomes&genomes * These rules are used to decide whether variants in genes with AD/AR inheritance should be reported in “Primary” or “Secondary” results in the clinical report (“Primary”: any P/LP variant in genes associated with AD or AR diseases, or VUS in AD genes; “Secondary”: any VUS in AR genes or VUS in AD/AR genes with ≥5 heterozygotes in gnomAD). |
| BS3 Well-established in vitro or in vivo functional studies show no damaging effect on protein function or splicing | VARIABLE | Consult PUBMED, LOVD and other available databases (Mastermind, LitVar, etc.) to find functional evidence [13]. |
| BS4 Lack of segregation in affected members of a family | NO | Segregation analysis required. |
| BP1 Missense variant in a gene for which primarily truncating variants are known to cause disease | YES (Mutually exclusive vs. PP2) |
Use SUPPORTING as default. |
| BP2 Observed in trans with a pathogenic variant for a fully penetrant dominant gene/disorder or observed in cis with a pathogenic variant in any inheritance pattern | NO | Segregation analysis required, use SUPPORTING as default. Use STRONG if the condition is confirmed in many individuals (literature or internal evidence) or with different variants. |
| BP3 In-frame deletions/insertions in a repetitive region without a known function | YES | Use SUPPORTING as default. |
| BP4 Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.) | YES | - Prediction must be based on at least 3 total and concordant predictors, otherwise exclude the criterion. - Do not assign to any variant with PVS1 activated. - The criterion must not be assigned to canonical splicing variants (±1–2) if PVS1 is assigned. If available, ADA and RF scores § can be used to assign the criterion to intronic variants. - Do not use ADA and RF scores to assign the criterion to synonymous variants if BP7 is already assigned. |
| BP5 Variant found in a case with an alternative molecular basis for disease | NO | Segregation analysis or literature evidence required. Use SUPPORTING as default. |
| BP6 Reputable source recently reports variant as benign, but the evidence is not available to the laboratory to perform an independent evaluation | YES | Exclude the criterion if the variant has “Review status” 0 stars in ClinVar and there are no other submissions in other clinical databases (e.g., LOVD). Be aware that certain UniProt classifications might be outdated. |
| BP7 A synonymous (silent) variant for which splicing prediction algorithms predict no impact on the splice consensus sequence nor creation of a new splice site AND the nucleotide is not highly conserved | YES | Use SUPPORTING as default. |
| PATHOGENIC CRITERIA | ||
| ACMG CRITERIA | ASSIGNED BY VARSOME | EXCEPTIONS AND INDICATIONS |
| PVS1 Null variant (nonsense, frameshift, canonical ±1 or 2 splice sites, initiation codon, single or multiexon deletion) in a gene where LoF is a known mechanism of disease | YES | - Modify the criterion strength according to Abu Tayoun et al. 2018 [10]. - the criterion does not apply to variants in the first/last base of an exon (not considered canonical in the ACMG guidelines). - use SUPPORTING if NMD is not predicted (variant in the last exon or in the last 50 bps of the second-last exon) AND there are no other P/LP variants downstream. |
| PS1 Same amino acid change as a previously established pathogenic variant regardless of nucleotide change | YES | Use STRONG as default, reduce to SUPPORTING if the alternative variant is classified as likely pathogenic. Always check interpretation of alternative variant. |
| PS2: De novo (confirmed maternity and paternity) in a patient with the disease and no family history | NO | Segregation analysis required. |
| PS3: Well-established in vitro or in vivo functional studies supporting a damaging effect on the gene or gene product | VARIABLE | - Consult PUBMED, LOVD, and other available databases (Mastermind, LitVar, etc.) to identify functional evidence - Modify criterion strength according to evidence relevance [13] |
| PS4 The prevalence of the variant in affected individuals is significantly higher than in controls | NO | - Use the criterion at STRONG level if prevalence data (cases/controls) are available - Use the criterion when the variant has been reported in at least 5 unrelated affected individuals in the laboratory |
| PM1: Located in a mutational hotspot and/or critical and well-established functional domain (e.g., the active site of an enzyme) without benign variation | YES | - Use MODERATE as default. - Reduce to SUPPORTING with <10 variants in the domain |
| PM2: Absent in controls (or extremely low frequency if recessive) in Exome Sequencing Project, 1000 Genomes Project, or Exome Aggregation Consortium | YES | Use MODERATE as default Since VarSome employs the CGD database, in case of known inheritance discrepancies, use the following gnomAD cutoffs * to include other variants in the selection: - AR/XL model: <3 homozygotes/hemizygotes in gnomAD exomes&genomes - AD model: <5 heterozygotes in gnomAD exomes&genomes * These rules are used to decide if variants in genes with AD/AR inheritance should be reported in “Primary” or “Secondary” results (“Primary”: any P/LP variant in genes associated with AD or AR disease, or VUS in AD genes; “Secondary”: any VUS in AR genes, or VUS in AD/AR genes with ≥5 heterozygotes in gnomAD). |
| PM3 For recessive disorders, detected in trans with a pathogenic variant | NO | - Use MODERATE as default - Use SUPPORTING if variant found in trans with only one other LP/P variant in one affected individual or for homozygous genotypes [29] - Upgrade to STRONG if found in trans with multiple different pathogenic variants or in multiple affected individuals (in the literature or in using internal segregation evidence). |
| PM4 Protein length changes as a result of in-frame deletions/insertions in a non-repeat region or stop-loss variants | YES (not applicable if PVS1 is enabled) |
Use MODERATE as default. |
| PM5: Novel missense change at an amino acid residue where a different missense change determined to be pathogenic has been seen before. | YES | Reduce to SUPPORTING in dubious cases, be aware that certain UniProt classification might be outdated, therefore always check interpretation of different missense changes. |
| PM6: Assumed de novo, but without confirmation of paternity and maternity. | NO | It is possible to modify criterion strength according to the compatibility of the proband’s phenotype with the disease associated with the gene and if the variant has been found de novo in other non-consanguineous individuals in the internal database. (Further implementations ongoing to refine grading). |
| PP1: Cosegregation with disease in multiple affected family members in a gene definitively known to cause the disease. | NO | Use: - STRONG: ≥4 total affected persons including the one tested - MODERATE: 2–3 affected persons including the tested individual - SUPPORTING: 1 affected person including the tested individual |
| PP2: Missense variant in a gene that has a low rate of benign missense variation and in which missense variants are a common mechanism of disease. | YES | Use SUPPORTING as default. |
| PP3: Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.) | YES | - Prediction must be based on at least 3 total and concordant predictors, otherwise exclude the criterion. - Do not assign to any variant with PVS1 activated. - The criterion must not be assigned to canonical splicing variants (±1–2) if PVS1 is assigned. If available, ADA and RF scores§ can be used to assign the criterion to intronic variants. |
| PP4: Patient’s phenotype or family history is highly specific for a disease with a single genetic etiology. | NO | - Use SUPPORTING for diseases with no more than 5 associated genes (e.g., Stargardt disease) - Use MODERATE for true single-gene disorders (e.g., CHM for choroideremia). |
| PP5: Reputable source recently reports variant as pathogenic, but the evidence is not available to the laboratory to perform an independent evaluation. | YES | Exclude the criterion if the variant has “Review status” 0 stars in ClinVar and there are no other submissions in other clinical databases (e.g., LOVD). Be aware that certain UniProt classifications might be outdated. |
§ Cutoffs: ADA score > 0.708; RF score > 0.515.