Heuristic methods for finding pathogenic variants in gene coding sequences

Differential Domain Distribution of gnomAD- and Disease-Linked Connexin Missense Variants

Twenty-one human genes encode connexins, a family of homologous proteins making gap junction (GJ) channels, which mediate direct intercellular communication to synchronize tissue/organ activities. Genetic variants in more than half of the connexin genes are associated with dozens of different Mendelian inherited diseases. With rapid advances in DNA sequencing technology, more variants are being identified not only in families and individuals with diseases but also in people in the general population without any apparent linkage to Mendelian inherited diseases. Nevertheless, it remains challenging to classify the pathogenicity of a newly identified connexin variant. Here, we analyzed the disease- and Genome Aggregation Database (gnomAD, as a proxy of the general population)-linked variants in the coding region of the four disease-linked α connexin genes. We found that the most abundant and position-sensitive missense variants showed distinct domain distribution preference between disease- and gnomAD-linked variants. Plotting missense variants on topological and structural models revealed that disease-linked missense variants are highly enriched on the structurally stable/resolved domains, especially the pore-lining domains, while the gnomAD-linked missense variants are highly enriched in the structurally unstable/unresolved domains, especially the carboxyl terminus. In addition, disease-linked variants tend to be on highly conserved residues and those positions show evolutionary co-variation, while the gnomAD-linked missense variants are likely on less conserved residue positions and on positions without co-variation. Collectively, the revealed distribution patterns of disease- and gnomAD-linked missense variants further our understanding of the GJ structure–biological function relationship, which is valuable for classifying the pathogenicity of newly identified connexin variants.

Predicting the Functional Impact of KCNQ1 Variants of Unknown Significance

BACKGROUND

An emerging standard-of-care for long-QT syndrome uses clinical genetic testing to identify genetic variants of the KCNQ1 potassium channel. However, interpreting results from genetic testing is confounded by the presence of variants of unknown significance for which there is inadequate evidence of pathogenicity.

METHODS AND RESULTS

In this study, we curated from the literature a high-quality set of 107 functionally characterized KCNQ1 variants. Based on this data set, we completed a detailed quantitative analysis on the sequence conservation patterns of subdomains of KCNQ1 and the distribution of pathogenic variants therein. We found that conserved subdomains generally are critical for channel function and are enriched with dysfunctional variants. Using this experimentally validated data set, we trained a neural network, designated Q1VarPred, specifically for predicting the functional impact of KCNQ1 variants of unknown significance. The estimated predictive performance of Q1VarPred in terms of Matthew's correlation coefficient and area under the receiver operating characteristic curve were 0.581 and 0.884, respectively, superior to the performance of 8 previous methods tested in parallel. Q1VarPred is publicly available as a web server at http://meilerlab.org/q1varpred.

CONCLUSIONS

Although a plethora of tools are available for making pathogenicity predictions over a genome-wide scale, previous tools fail to perform in a robust manner when applied to KCNQ1. The contrasting and favorable results for Q1VarPred suggest a promising approach, where a machine-learning algorithm is tailored to a specific protein target and trained with a functionally validated data set to calibrate informatics tools.

Development of pathogenicity predictors specific for variants that do not comply with clinical guidelines for the use of computational evidence

Background

Strict guidelines delimit the use of computational information in the clinical setting, due to the still moderate accuracy of in silico tools. These guidelines indicate that several tools should always be used and that full coincidence between them is required if we want to consider their results as supporting evidence in medical decision processes. Application of this simple rule certainly decreases the error rate of in silico pathogenicity assignments. However, when predictors disagree this rule results in the rejection of potentially valuable information for a number of variants. In this work, we focus on these variants of the protein sequence and develop specific predictors to help improve the success rate of their annotation.

Results

We have used a set of 59,442 protein sequence variants (15,723 pathological and 43,719 neutral) from 228 proteins to identify those cases for which pathogenicity predictors disagree. We have repeated this process for all the possible combinations of five known methods (SIFT, PolyPhen-2, PON-P2, CADD and MutationTaster2). For each resulting subset we have trained a specific pathogenicity predictor. We find that these specific predictors are able to discriminate between neutral and pathogenic variants, with a success rate different from random. They tend to outperform the constitutive methods but this trend decreases as the performance of the constitutive predictor improves (e.g. with PON-P2 and PolyPhen-2). We also find that specific methods outperform standard consensus methods (Condel and CAROL).

Conclusion

Focusing development efforts on the case of variants for which known methods disagree we may obtain pathogenicity predictors with improved performances. Although we have not yet reached the success rate that allows the use of this computational evidence in a clinical setting, the simplicity of the approach indicates that more advanced methods may reach this goal in a close future.

Electronic supplementary material

The online version of this article (doi:10.1186/s12864-017-3914-0) contains supplementary material, which is available to authorized users.

Genetic coding variants in the niacin receptor, hydroxyl-carboxylic acid receptor 2, and response to niacin therapy

OBJECTIVE

Niacin has been used for seven decades to modulate plasma lipids, but its mechanism of action is still unclear. We sought to determine whether variants in the niacin receptor gene, hydroxyl-carboxylic receptor 2 (HCAR2), are associated with lipid response to treatment.

PARTICIPANTS AND METHODS

Coding variants, rs7314976 (p.R311C) and rs2454727 (p.M317I), were genotyped in 2067 participants from the Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides and Impact on Global Health Outcomes (AIM-HIGH) trial. AIM-HIGH was a randomized, placebo-controlled trial that was conducted to assess the effect of extended-release niacin in patients with cardiovascular disease aggressively treated with low-density lipoprotein cholesterol-lowering therapy.

RESULTS

There was no association of p.R311C or p.M317I with changes in low-density lipoprotein cholesterol, triglycerides, or high-density lipoprotein cholesterol at 1 year in groups receiving placebo or extended-release niacin. In White patients, the reduction in lipoprotein (a) [Lp(a)] in response to niacin was greater in homozygous carriers of the major 317M allele (-22.7%; P=0.005) compared with minor allele carriers (-15.3%). This was directionally consistent in the Black participants. Upon combining both groups, the reduction in Lp(a) in response to niacin was significantly greater in the homozygous major allele carriers (-23.0%; P=0.003) compared with minor allele carriers (-15.2%).

CONCLUSION

Understanding the genetic contribution toward variation in response to niacin therapy, including Lp(a) reduction, could uncover mechanisms by which niacin decreases Lp(a), an important independent risk factor for cardiovascular disease.

The Role of Pharmacogenetics in Atrial Fibrillation Therapeutics: Is Personalized Therapy in Sight?

Atrial fibrillation (AF) is the most common sustained cardiac arrhythmia worldwide requiring therapy. Despite recent advances in catheter-based and surgical therapy, antiarrhythmic drugs (AADs) remain the mainstay of treatment for symptomatic AF. However, response in individual patients is highly variable with over half the patients treated with rhythm control therapy experiencing recurrence of AF within a year. Contemporary AADs used to suppress AF are incompletely and unpredictably effective and associated with significant risks of proarrhythmia and noncardiac toxicities. Furthermore, this "one-size" fits all strategy for selecting antiarrhythmics is based largely on minimizing risk of adverse effects rather than on the likelihood of suppressing AF. The limited success of rhythm control therapy is in part due to heterogeneity of the underlying substrate, interindividual differences in disease mechanisms, and our inability to predict response to AADs in individual patients. Genetic studies of AF over the past decade have revealed that susceptibility to and response to therapy for AF is modulated by the underlying genetic substrate. However, the bedside application of these new discoveries to the management of AF patients has thus far been disappointing. This may in part be related to our limited understanding about genetic predictors of drug response in general, the challenges associated with determining efficacy of response to AADs, and lack of randomized genotype-directed clinical trials. Nonetheless, recent studies have shown that common AF susceptibility risk alleles at the chromosome 4q25 locus modulated response to AADs, electrical cardioversion, and ablation therapy. This monograph discusses how genetic approaches to AF have not only provided important insights into underlying mechanisms but also identified AF subtypes that can be better targeted with more mechanism-based "personalized" therapy.

Molecular damage in Fabry disease: characterization and prediction of alpha-galactosidase A pathological mutations

Loss-of-function mutations of the enzyme alpha-galactosidase A (GLA) causes Fabry disease (FD), that is a rare and potentially fatal disease. Identification of these pathological mutations by sequencing is important because it allows an early treatment of the disease. However, before taking any treatment decision, if the mutation identified is unknown, we first need to establish if it is pathological or not. General bioinformatic tools (PolyPhen-2, SIFT, Condel, etc.) can be used for this purpose, but their performance is still limited. Here we present a new tool, specifically derived for the assessment of GLA mutations. We first compared mutations of this enzyme known to cause FD with neutral sequence variants, using several structure and sequence properties. Then, we used these properties to develop a family of prediction methods adapted to different quality requirements. Trained and tested on a set of known Fabry mutations, our methods have a performance (Matthews correlation: 0.56-0.72) comparable or better than that of the more complex method, Polyphen-2 (Matthews correlation: 0.61), and better than those of SIFT (Matthews correl.: 0.54) and Condel (Matthews correl.: 0.51). This result is validated in an independent set of 65 pathological mutations, for which our method displayed the best success rate (91.0%, 87.7%, and 73.8%, for our method, PolyPhen-2 and SIFT, respectively). These data confirmed that our specific approach can effectively contribute to the identification of pathological mutations in GLA, and therefore enhance the use of sequence information in the identification of undiagnosed Fabry patients.

Aryl hydrocarbon receptor interacting protein (AIP) mutations occur rarely in sporadic parathyroid adenomas

CONTEXT

The molecular pathogenesis of primary hyperparathyroidism is still largely unknown. The aryl hydrocarbon receptor interacting protein (AIP) gene has a major role in the pathogenesis of familial isolated pituitary adenoma.

OBJECTIVE

We evaluated the involvement of the AIP gene in sporadic parathyroid adenomas.

PATIENTS AND DESIGN

We performed direct sequencing and multiplex ligation-dependent probe amplification analyses of the AIP gene in a large series of sporadic parathyroid adenomas. Loss of heterozygosity (LOH) at the AIP locus was studied, and aryl hydrocarbon receptor interacting protein immunostaining was also performed. In addition, alterations in the MEN1 gene were studied.

RESULTS

A somatic AIP mutation, substitution of arginine with glutamine at codon 304 (R304Q), was identified in 2 of 132 tumors. The mutation was germline in both cases despite the nonfamilial presentation. Heterozygous AIP large deletions were detected in 29 cases including 1 of the 2 mutated tumors, confirming a biallelic inactivation of the AIP gene. The AIP-mutated tumor with LOH showed decreased AIP immunostaining compared with normal parathyroid. LOH at the MEN1 locus, which often shared LOH at the AIP locus, was found in one third of tumors. Somatic MEN1 mutations were found in the 1 of the 2 AIP-mutated tumors and in 22 parathyroid adenomas. In addition, multiplex ligation-dependent probe amplification analysis revealed 1 large deletion of the MEN1 gene in 1 patient.

CONCLUSIONS

The AIP gene is rarely involved in parathyroid adenomas, but the germline nature of the mutations suggests that it might predispose to primary hyperparathyroidism. MEN1 gene alterations occur in a substantial proportion of sporadic parathyroid adenomas.