Table 1a.

Evidence used to classify variants according to their pathogenicity level

Evidence level	Criteria
Pathogenic Strong	1) Coding amino-acid change previously published as deleterious with evidence of segregation in more than one pedigree or in multiple unrelated patients with the same phenotype
	2) Null variant in a gene where loss of function (LOF) is a known disease mechanism (caveat LOF variants at extreme 3’-end)
	3) Variant in a gene associated with an expected very rare pathology (e.g., PRNP mutation and prion pathology)
	4) Explained mechanism of pathophysiology of variant using in vitro or in vivo studies
	5) Found in a mutational hotspot, i.e., a domain where many other pathogenic mutations are seen, generally with additionally support from in silico prediction software
Pathogenic Moderate	1) Coding amino-acid change previously and justifiably published as deleterious but without evidence of segregation or in a single pedigree/patient
	2) Novel missense change at an amino-acid residue where a different pathogenic missense change has been seen
	3) A very different amino-acid change at the same site or next to one with a less dramatic amino-acid change but deleterious
	4) In a gene, the mechanism of which is understood and the effect of the variant is in keeping with that mechanism
	5) Protein length changes as a result of in-frame deletions/insertions in a nonrepeat region or stop-loss variants
	6) Mutation in a gene associated with a rare pathology in a case with a compatible clinical syndrome
	7) Intronic variant affecting splicing or protein length
Pathogenic Supporting	1) Variant with a major amino-acid change near or in a functional domain (e.g., active site of an enzyme) but not in a mutational hotspot
	2) Multiple lines of computational evidence support a deleterious effect on the gene or gene product (conservation, evolutionary, splicing impact, etc.), caveat: because many in silico algorithms use the same or very similar input for their predictions, each algorithm should not be counted as an independent criterion
	3) Reported in both cases and controls, but more cases than controls (statistically significant in a study)
Pathogenic Risk Factor	1) Previously reported as risk factor, either variant itself or clear established pattern in gene
	2) > 1 in 10000 in gnomAD
	3) The prevalence of the variant in affected individuals is significantly increased compared with the prevalence in controls
Benign Independent	Allele frequency > 5% on gnomAD, or 1000 genomes project
Benign Strong	1) Allele frequency > 1% on gnomAD
	2) Reported benign in multiple pedigrees or with insight into gene/protein mechanism
	3) Allele frequency is greater than expected for disorder
	4) Lack of segregation in affected members of a family, caveat: phenocopies and penetrance
	5) Seen in equal or greater frequencies in controls than cases
Benign Moderate	1) Allele frequency over 0.1% on gnomAD
	2) Reported benign in one case or pedigree
	3) Genetic mechanism inconsistent with pathological phenotype, or known mutation spectrum
Benign Supporting	1) Missense variant in a gene for which primarily truncating variants are known to cause disease or the mechanism is very specific and known
	2) Multiple lines of computational evidence suggest no impact on gene or gene product (conservation, evolutionary, splicing impact, etc.)
	3) A synonymous (silent) variant for which splicing prediction algorithms predict no impact to the splice consensus sequence

Variants identified in a sample were classified according to the information available about them. This included the type of mutation in question, its position in the gene and/or protein, its frequency in online population databases, in silico predictions of effects on proteins, and whether it had previously been reported in families, single cases or controls