Identification and in-silico characterization of splice-site variants from a large cardiogenetic national registry

Abstract

Splice-site variants in cardiac genes may predispose carriers to potentially lethal arrhythmias. To investigate, we screened 1315 probands and first-degree relatives enrolled in the Canadian Hearts in Rhythm Organization (HiRO) registry. 10% (134/1315) of patients in the HiRO registry carry variants within 10 base-pairs of the intron-exon boundary with 78% (104/134) otherwise genotype negative. These 134 probands were carriers of 57 unique variants. For each variant, American College of Medical Genetics and Genomics (ACMG) classification was revisited based on consensus between nine in silico tools. Due in part to the in silico algorithms, seven variants were reclassified from the original report, with the majority (6/7) downgraded. Our analyses predicted 53% (30/57) of variants to be likely/pathogenic. For the 57 variants, an average of 9 tools were able to score variants within splice sites, while 6.5 tools responded for variants outside these sites. With likely/pathogenic classification considered a positive outcome, the ACMG classification was used to calculate sensitivity/specificity of each tool. Among these, Combined Annotation Dependent Depletion (CADD) had good sensitivity (93%) and the highest response rate (131/134, 98%), dbscSNV was also sensitive (97%), and SpliceAI was the most specific (64%) tool. Splice variants remain an important consideration in gene elusive inherited arrhythmia syndromes. Screening for intronic variants, even when restricted to the ±10 positions as performed here may improve genetic testing yield. We compare 9 freely available in silico tools and provide recommendations regarding their predictive capabilities. Moreover, we highlight several novel cardiomyopathy-associated variants which merit further study.

Fig. 1. RNA splicing schematic.

The splice donor/acceptor sites, splicing of the intronic lariat, and protein product are labelled.

Fig. 2. Work-flow of utilized in silico tools.

The in silico tools and other online tools used in the analysis of splice-site variants are outlined in a stepwise manner. Deleterious Annotation Using Neural Networks (DANN) evaluates genome wide, MutationTaster evaluates genome wide but test data sets were limited to donor (last 3 exonic bases and 6 first intronic bases) and acceptor (last 12 intronic bases and 2 first exonic bases) splice sites., dbscSNV (ADA and RF score) evaluates from −3→ +8 at the 5ʹ and −12→+2 at the 3ʹ end, Splice Site Prediction by Neural Network (NNSplice), Combined Annotation Dependent Depletion (CADD) evaluates genome wide, SpliceAI evaluates genome wide, Human Splice Finder (HSF) evaluates genome wide, MaxEntScan (MES) evaluates −3→+6 at 5’ end and −20→+3 at 3ʹ end.

Fig. 3. Position of identified splice-site variants.

This figure indicates the position of variants relative to the splice donor/acceptor site as a percentage of patient carriers. The label on each column indicates the number of individuals found to carry a variant within the indicated distance from the splice site. 8% of the variants were missense. 57% of variants occurred in the ±2, 85% in ±4, 88% in ±6, 90% in ±8, and 93% in ±10 positions.

Fig. 4. Predictive capability of in silico tools.

The sensitivity (white) and specificity (black) for each tool is displayed as a percentage. Sensitivity was determined from the number of true positives as a ratio of true positives and false negatives. Specificity was determined from the number of true negatives as a ratio of true negatives and false positives. The final classification of LP or P was considered a positive result and any other ACMG-AMP classification considered negative for each prediction. The table lists the false negatives, false positives, true negatives, and true positives that were used to calculate the sensitivity and specificity for each tool.