Evaluation of Bioinformatic Programmes for the Analysis of Variants within Splice Site Consensus Regions

Abstract

The increasing diagnostic use of gene sequencing has led to an expanding dataset of novel variants that lie within consensus splice junctions. The challenge for diagnostic laboratories is the evaluation of these variants in order to determine if they affect splicing or are merely benign. A common evaluation strategy is to use in silico analysis, and it is here that a number of programmes are available online; however, currently, there are no consensus guidelines on the selection of programmes or protocols to interpret the prediction results. Using a collection of 222 pathogenic mutations and 50 benign polymorphisms, we evaluated the sensitivity and specificity of four in silico programmes in predicting the effect of each variant on splicing. The programmes comprised Human Splice Finder (HSF), Max Entropy Scan (MES), NNSplice, and ASSP. The MES and ASSP programmes gave the highest performance based on Receiver Operator Curve analysis, with an optimal cut-off of score reduction of 10%. The study also showed that the sensitivity of prediction is affected by the level of conservation of individual positions, with in silico predictions for variants at positions −4 and +7 within consensus splice sites being largely uninformative.

Thorough in silico and in vitro cDNA analysis of 21 putativeBRCA1andBRCA2splice variants and a complex tandem duplication inBRCA2allowing the identification of activated cryptic splice donor sites inBRCA2exon 11

“…Table S1 and Figure 4A). MES provided a small reduction in score (−12.1%, lower than the 15% cut-off previously proposed by Houdayer et al (2012) and Jian et al (2014a), but higher than the 10% cut-off more recently proposed by Tang et al (2016)). Priors calculated a greater change of pathogenicity through splicing defects (probability = 0.34) than as to be expected from the F I G U R E 1 Results of in vitro mRNA analysis of six BRCA1 variants leading to aberrant splicing.…”

“…(), but higher than the 10% cut‐off more recently proposed by Tang et al. ()). Priors calculated a greater change of pathogenicity through splicing defects (probability = 0.34) than as to be expected from the missense variant itself (probability = 0.02).…”

“…Based on the MES predictions, aberrant splicing for BRCA1 c.4186‐10G > A (−17.7%) and BRCA2 c.8488‐12A > G (−13.7%) could be expected based on the −10% cut‐off proposed by Tang et al. (), whereas Houdayer et al. () applied a more stringent cut‐off of −15%.…”

“…Indeed, variants in the highly conserved dinucleotides at the splice site donor/acceptor sites at the intron boundaries, in the conserved adjacent sequences or variants in the branchpoints and exonic splicing enhancers (ESE) might lead to aberrant splicing (Houdayer et al, 2012;Spurdle, Couch, Hogervorst, Radice, & Sinilnikova, 2008;Strachan & Read, 2010). Several guidelines have been formulated to assist the decision-making process (Caminsky, Mucaki, & Rogan, 2014;Houdayer et al, 2008;Tang, Prosser, & Love, 2016). Accurate prediction by these tools is dependent on the ability to detect the wild-type (WT) splice site or define regions such as the branchpoints or ESE.…”

Splice Site Variants in the KCNQ1 and SCN5A Genes: Transcript Analysis as a Tool in Supporting Pathogenicity

“…These programmes predict whether a splice-site variant is potentially pathogenic, but their efficacy varies [12]. The ideal test of whether a variant affects splicing is transcript analysis using RNA isolated from the affected tissue as cis- acting splicing mutations can have cell-specific effects [10].…”