Clinical likelihood ratios and balanced accuracy for 44 in silico tools against multiple large-scale functional assays of cancer susceptibility genes.

PURPOSE: Where multiple in silico tools are concordant, the American College of Medical Genetics and Genomics/Association for Molecular Pathology (ACMG/AMP) framework affords supporting evidence toward pathogenicity or benignity, equivalent to a likelihood ratio of ~2. However, limited availability of "clinical truth sets" and prior use in tool training limits their utility for evaluation of tool performance. METHODS: We created a truth set of 9,436 missense variants classified as deleterious or tolerated in clinically validated high-throughput functional assays for BRCA1, BRCA2, MSH2, PTEN, and TP53 to evaluate predictive performance for 44 recommended/commonly used in silico tools. RESULTS: Over two-thirds of the tool-threshold combinations examined had specificity of <50%, thus substantially overcalling deleteriousness. REVEL scores of 0.8-1.0 had a Positive Likelihood Ratio (PLR) of 6.74 (5.24-8.82) compared to scores <0.7 and scores of 0-0.4 had a Negative Likelihood Ratio (NLR) of 34.3 (31.5-37.3) compared to scores of >0.7. For Meta-SNP, the equivalent PLR = 42.9 (14.4-406) and NLR = 19.4 (15.6-24.9). CONCLUSION: Against these clinically validated "functional truth sets," there was wide variation in the predictive performance of commonly used in silico tools. Overall, REVEL and Meta-SNP had best balanced accuracy and might potentially be used at stronger evidence weighting than current ACMG/AMP prescription, in particular for predictions of benignity.

Glycogen storage disease type III in the Irish population.

Glycogen storage disease type III (GSD III) results from mutations of the AGL gene encoding the glycogen debrancher enzyme. The disease has clinical and biochemical heterogeneity reflecting the severity of the AGL mutations. We sought to characterise the molecular defects in our cohort of Irish patients with GSD III. Fifteen patients from eight unrelated Irish families were identified: six males and nine females. The age ranged from 2-39 years old, and all presented in the first 3 years of life. Four patients (of three families) had mild disease with hepatomegaly, mild hypoglycaemia and normal creatine kinase (CK) levels. Five families had more severe disease, with liver and skeletal muscle involvement and elevated CK. Eleven different mutations were identified amongst the eight families. Of the 11, six were novel: p.T512fs, p.S736fs, p.A1400fs, p.K1407fs, p.Y519X and p.D627Y. The family homozygous for p.A1400fs had the most severe phenotype (early-onset hypoglycaemia, massive hepatomegaly, myopathy and hypertrophic cardiomyopathy before age 2 years), which was not halted by aggressive carbohydrate and protein supplementation. Conversely, the only missense mutation identified in the cohort, p.D627Y, was associated with a mild phenotype. The phenotypic diversity in our GSD III cohort is mirrored by the allelic heterogeneity. We describe two novel null mutations in exon 32 in two families with severe GSD III resistant to current treatment modalities. Knowledge of the specific mutations segregating in this cohort may allow for the development of new therapeutic interventions.

A study of Wilson disease mutations in Britain.

Wilson disease (WD) is an autosomal recessive disease of copper transport. The disease is caused by a large number of mutations in the ATP7B gene, some of which appear to be population specific, whereas others are found in probands from a variety of different ethnic backgrounds. This study presents the results of screening the ATP7B gene by SSCP and sequencing in order to define the spectrum of mutations seen in British referrals for WD. The 52 patients screened included 10 with a non-British mixed ethnicity origin. This study identified 19 novel mutations and 18 mutations that had been previously described. The novel mutations included seven nonconservative missense mutations, eight small insertions, or deletions causing frameshift, two nonsense mutations, and two splice-site mutations. Seven of the 10 mixed ethnicity patients harboured homozygous mutations, whereas only four of the larger British group were homozygotes. The detection rate by SSCP analysis in the British group of 42 consecutive unrelated WD probands was 70%. However, SSCP screening of just three exons (exons 8, 14, and 18) is predicted to identify 60% of mutations present in WD referrals.

Maternal sex chromosome non-disjunction: evidence for X chromosome-specific risk factors.

Human trisomy is attributable to many different mechanisms and the relative importance of each mechanism is highly chromosome specific. The association between altered recombination and maternal non-disjunction is well documented: reductions in recombination have been reported for maternal meiosis I (MI) errors involving chromosomes 15, 16, 18 and 21 and increased recombination has been reported for meiosis II (MII) errors involving chromosome 21. We therefore investigated maternal X chromosome non-disjunction, to determine whether the effects of recombination are unique to the X chromosome or similar to any of the autosomes thus far studied. We genotyped 45 47,XXX females and 95 47,XXY males of maternal origin. Our results demonstrate that 49% arose during MI, 29% during MII and 16% were postzygotic events; a further 7% were meiotic but could not be assigned as either MI or MII because of recombination at the centromere. Among the MI cases, a majority (56%) had no detectable transitions and so absent recombination is an important factor for X chromosome non-disjunction. However, similar to trisomy 15 and unlike trisomy 21, we observed a significant increase in the mean maternal age of transitional MI errors compared with nullitransitional cases. In our studies of MII errors, recombination appeared normal and there was no obvious effect of maternal age, distinguishing our results from MII non-disjunction of chromosomes 18 or 21. Thus, surprisingly, the risk factors associated with both MI and MII non-disjunction appear to be different for virtually every chromosome that has been adequately studied.