Marc S. Greenblatt

Purpose: We previously developed an approach to calibrate computational tools for clinical variant classification, updating recommendations for the reliable use of variant impact predictors to provide evidence strength up to Strong . A new generation of tools using distinctive approaches have since been released, and these methods must be independently calibrated for clinical application. Method: Using our local posterior probability-based calibration and our established data set of ClinVar pathogenic and benign variants, we determined the strength of evidence provided by three new tools (AlphaMissense, ESM1b, VARITY) and calibrated scores meeting each evidence strength. Results All three tools reached the Strong level of evidence for variant pathogenicity and Moderate for benignity, though sometimes for few variants. Compared to previously recommended tools, these yielded at best only modest improvements in the tradeoffs of evidence strength and false positive predictions. Conclusion: At calibrated thresholds, three new computational predictors provided evidence for variant pathogenicity at similar strength to the four previously recommended predictors (and comparable with functional assays for some variants). This calibration broadens the scope of computational tools for application in clinical variant classification. Their new approaches offer promise for future advancement of the field.

Purpose To investigate the number of rare missense variants observed in human genome sequences by ACMG/AMP PP3/BP4 evidence strength, following the calibrated PP3/BP4 computational recommendations. Methods Missense variants from the genome sequences of 300 probands from the Rare Genomes Project with suspected rare disease were analyzed using computational prediction tools able to reach PP3_Strong and BP4_Moderate evidence strengths (BayesDel, MutPred2, REVEL, and VEST4). The numbers of variants at each evidence strength were analyzed across disease-associated genes and genome-wide. Results From a median of 75.5 rare (≤1% allele frequency) missense variants in disease-associated genes per proband, a median of one reached PP3_Strong, 3-5 PP3_Moderate, and 3-5 PP3_Supporting. Most were allocated BP4 evidence (median 41-49 per proband) or were indeterminate (median 17.5-19 per proband). Extending the analysis to all protein-coding genes genome-wide, the number of PP3_Strong variants increased approximately 2.6-fold compared to disease-associated genes, with a median per proband of 1-3 PP3_Strong, 8-16 PP3_Moderate, and 10-17 PP3_Supporting. Conclusion A small number of variants per proband reached PP3_Strong and PP3_Moderate in 3,424 disease-associated genes, and though not the intended use of the recommendations, also genome-wide. Use of PP3/BP4 evidence as recommended from calibrated computational prediction tools in the clinical diagnostic laboratory is unlikely to inappropriately contribute to the classification of an excessive number of variants as Pathogenic or Likely Pathogenic by ACMG/AMP rules.

MUTYH‐associated polyposis (MAP) is an autosomal recessive disorder where the inheritance of constitutional biallelic pathogenic MUTYH variants predisposes a person to the development of adenomas and colorectal cancer (CRC). It is also associated with extracolonic and extraintestinal manifestations that may overlap with the phenotype of familial adenomatous polyposis (FAP). Currently, there are discrepancies in the literature regarding whether certain phenotypes are truly associated with MAP. This narrative review is aimed at exploring the phenotypic spectrum of MAP to better characterize the MAP phenotype. Literature search was conducted to identify articles reporting on MAP‐specific phenotypes. Clinical data from 2109 MAP patients identified from the literature showed that 1123 patients (53.2%) had CRC. Some patients with CRC had no associated adenomas, suggesting that adenomas are not an obligatory component of MAP. Carriers of the two missense founder variants, and possibly truncating variants, had an increased cancer risk when compared to those who carry other pathogenic variants. It has been suggested that somatic G:C > T:A transversions are a mutational signature of MAP and could be used as a biomarker in screening and identifying patients with atypical MAP, or in associating certain phenotypes with MAP. The extracolonic and extraintestinal manifestations that have been associated with MAP include duodenal adenomas, duodenal cancer, fundic gland polyps, gastric cancer, ovarian cancer, bladder cancer, and skin cancer. The association of breast cancer and endometrial cancer with MAP remains disputed. Desmoid tumors and congenital hypertrophy of the retinal pigment epithelium (CHRPEs) are rarely reported in MAP but have long been seen in FAP patients and thus could act as a distinguishing feature between the two. This collection of MAP phenotypes will assist in the assessment of pathogenic MUTYH variants using the American College of Medical Genetics and the Association for Molecular Pathology (ACMG/AMP) Variant Interpretation Guidelines and ultimately improve patient care.

Recommendations from the American College of Medical Genetics and Genomics and the Association for Molecular Pathology (ACMG/AMP) for interpreting sequence variants specify the use of computational predictors as Supporting level of evidence for pathogenicity or benignity using criteria PP3 and BP4, respectively. However, score intervals defined by tool developers, and ACMG/AMP recommendations that require the consensus of multiple predictors, lack quantitative support. Previously, we described a probabilistic framework that quantified the strengths of evidence (Supporting, Moderate, Strong, Very Strong) within ACMG/AMP recommendations. We have extended this framework to computational predictors and introduce a new standard that converts a tool’s scores to PP3 and BP4 evidence strengths. Our approach is based on estimating the local positive predictive value and can calibrate any computational tool or other continuous-scale evidence on any variant type. We estimate thresholds (score intervals) corresponding to each strength of evidence for pathogenicity and benignity for thirteen missense variant interpretation tools, using carefully assembled independent data sets. Most tools achieved Supporting evidence level for both pathogenic and benign classification using newly established thresholds. Multiple tools reached score thresholds justifying Moderate and several reached Strong evidence levels. One tool reached Very Strong evidence level for benign classification on some variants. Based on these findings, we provide recommendations for evidence-based revisions of the PP3 and BP4 ACMG/AMP criteria using individual tools and future assessment of computational methods for clinical interpretation.

Background. Lynch syndrome is the most common genetic predisposition for hereditary cancer. Carriers of pathogenic changes in mismatch repair (MMR) genes have an increased risk of developing colorectal (CRC), endometrial, ovarian, urinary tract, prostate, and other cancers, depending on which gene is malfunctioning. In Lynch syndrome, differences in cancer incidence (penetrance) according to the gene involved have led to the stratification of cancer surveillance. By contrast, any differences in penetrance determined by the type of pathogenic variant remain unknown. Objective. To determine cumulative incidences of cancer in carriers of truncating and missense or aberrant splicing pathogenic variants of the MLH1 and MSH2 genes. Methods. Carriers of pathogenic variants of MLH1 (path_MLH1) and MSH2 (path_MSH2) genes filed in the Prospective Lynch Syndrome Database (PLSD) were categorized as truncating or missense/aberrant splicing according to the InSiGHT criteria for pathogenicity. Results. Among 5199 carriers, 1045 had missense or aberrant splicing variants, and 3930 had truncating variants. Prospective observation years for the two groups were 8205 and 34,141 years, respectively, after which there were no significant differences in incidences for cancer overall or for colorectal cancer or endometrial cancers separately. Conclusion. Truncating and missense or aberrant splicing pathogenic variants were associated with similar average cumulative incidences of cancer in carriers of path MLH1 and path_MSH2.

Computational algorithms are often used to assess pathogenicity of Variants of Uncertain Significance (VUS) that are found in disease-associated genes. Most computational methods include analysis of protein multiple sequence alignments (PMSA), assessing interspecies variation. Careful validation of PMSA-based methods has been done for relatively few genes, partially because creation of curated PMSAs is labor-intensive. We assessed how PMSA-based computational tools predict the effects of the missense changes in the APC gene, in which pathogenic variants cause Familial Adenomatous Polyposis. Most Pathogenic or Likely Pathogenic APC variants are protein-truncating changes. However, public databases now contain thousands of variants reported as missense. We created a curated APC PMSA that contained >3 substitutions/site, which is large enough for statistically robust in silico analysis. The creation of the PMSA was not easily automated, requiring significant querying and computational analysis of protein and genome sequences. Of 1924 missense APC variants in the NCBI ClinVar database, 1800 (93.5%) are reported as VUS. All but two missense variants listed as P/LP occur at canonical splice or Exonic Splice Enhancer sites. Pathogenicity predictions by five computational tools (Align-GVGD, SIFT, PolyPhen2, MAPP, REVEL) differed widely in their predictions of Pathogenic/Likely Pathogenic (range 17.5–75.0%) and Benign/Likely Benign (range 25.0–82.5%) for APC missense variants in ClinVar. When applied to 21 missense variants reported in ClinVar and securely classified as Benign, the five methods ranged in accuracy from 76.2–100%. Computational PMSA-based methods can be an excellent classifier for variants of some hereditary cancer genes. However, there may be characteristics of the APC gene and protein that confound the results of in silico algorithms. A systematic study of these features could greatly improve the automation of alignment-based techniques and the use of predictive algorithms in hereditary cancer genes.

Computational algorithms are often used to assess pathogenicity of Variants of Uncertain Significance (VUS) that are found in disease-associated genes. Most computational methods include analysis of protein multiple sequence alignments (PMSA), assessing interspecies variation. Careful validation of PMSA-based methods has been done for relatively few genes, partially because creation of curated PMSAs is labor-intensive. We assessed how PMSA-based computational tools predict the effects of the missense changes in the APC gene, in which pathogenic variants cause Familial Adenomatous Polyposis. Most Pathogenic or Likely Pathogenic APC variants are protein-truncating changes. However, public databases now contain thousands of variants reported as missense. We created a curated APC PMSA that contained >3 substitutions/site, which is large enough for statistically robust in silico analysis. The creation of the PMSA was not easily automated, requiring significant querying and computational analysis of protein and genome sequences. Of 1924 missense APC variants in the NCBI ClinVar database, 1800 (93.5%) are reported as VUS. All but two missense variants listed as P/LP occur at canonical splice or Exonic Splice Enhancer sites. Pathogenicity predictions by five computational tools (Align-GVGD, SIFT, PolyPhen2, MAPP, REVEL) differed widely in their predictions of Pathogenic/Likely Pathogenic (range 17.5–75.0%) and Benign/Likely Benign (range 25.0–82.5%) for APC missense variants in ClinVar. When applied to 21 missense variants reported in ClinVar as Benign, the five methods ranged in accuracy from 76.2-100%. Computational PMSA-based methods can be an excellent classifier for variants of some hereditary cancer genes. However, there may be characteristics of the APC gene and protein that confound the results of in silico algorithms. A systematic study of these features could greatly improve the automation of alignment-based techniques and the use of predictive algorithms in hereditary cancer genes. Author Summary: A critical problem in clinical genetics today is interpreting whether a genetic variant is benign or causes disease (pathogenic). Some of the hardest variants to interpret are those that change one amino acid for another in a protein sequence (a “missense variant”). Various computer programs are often used to predict whether mutations in disease-associated genes likely cause disease. Most computer programs involve studying how the gene has changed during evolution, comparing the protein sequences of different species by aligning them with each other. Variants in amino acids that have not tolerated mutation during evolution are usually predicted to be pathogenic, and variants in amino acids that have tolerated variation are usually predicted to be benign. High quality alignments are necessary to make accurate predictions. However, creating high quality alignments is difficult, not easily automated, and requires significant manual curation. Results from computer-generated predictions are used in current published guidelines as one tool for evaluating whether variants will disrupt the protein function and cause disease. These guidelines may be applied to genes in which single amino acid substitutions do not commonly cause disease. One such example is the APC gene, which is responsible for Familial Adenomatous Polyposis (FAP). Missense APC changes are not a common cause of FAP. Our analysis of APC demonstrated the difficulty of generating an accurate protein sequence alignment and the tendency of computer tools to overestimate the damaging effects of amino acid substitutions. Our results suggest that the rules for using computer-based tools to predict whether a variant causes disease should be modified when applied to genes in which missense variants rarely cause disease.

Background The American College of Medical Genetics and Genomics (ACMG)/Association for Molecular Pathology (AMP) clinical variant interpretation guidelines established criteria (PS3/BS3) for functional assays that specified a “strong” level of evidence. However, they did not provide detailed guidance on how functional evidence should be evaluated, and differences in the application of the PS3/BS3 codes is a contributor to variant interpretation discordance between laboratories. This recommendation seeks to provide a more structured approach to the assessment of functional assays for variant interpretation and guidance on the use of various levels of strength based on assay validation. Methods The Clinical Genome Resource (ClinGen) Sequence Variant Interpretation (SVI) Working Group used curated functional evidence from ClinGen Variant Curation Expert Panel-developed rule specifications and expert opinions to refine the PS3/BS3 criteria over multiple in-person and virtual meetings. We estimated odds of pathogenicity for assays using various numbers of variant controls to determine the minimum controls required to reach moderate level evidence. Feedback from the ClinGen Steering Committee and outside experts were incorporated into the recommendations at multiple stages of development. Results The SVI Working Group developed recommendations for evaluators regarding the assessment of the clinical validity of functional data and a four-step provisional framework to determine the appropriate strength of evidence that can be applied in clinical variant interpretation. These steps are: 1. Define the disease mechanism; 2. Evaluate applicability of general classes of assays used in the field; 3. Evaluate validity of specific instances of assays; 4. Apply evidence to individual variant interpretation. We found that a minimum of eleven total pathogenic and benign variant controls are required to reach moderate-level evidence in the absence of rigorous statistical analysis. Conclusions The recommendations and approach to functional evidence evaluation described here should help clarify the clinical variant interpretation process for functional assays. Further, we hope that these recommendations will help develop productive partnerships with basic scientists who have developed functional assays that are useful for interrogating the function of a variety of genes.