Ultra-rare variants in the TRPV6 gene cause autosomal dominant chronic pancreatitis

Masson, Emmanuelle; Chen, J.; Rebours, Vinciane; Rhun, Marc Le; Audin, H.; Lachaux, A.; Caumont, B.; Lorenzo, Diane; Fron, M.; Billiemaz, Kareen; Besnard, R.; Buscail, Louis; Férec, Claude

doi:10.1016/j.pan.2020.07.383

Cited by 1 publication

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recent years, there have been significant advancements in the prediction of splicing outcomes for SNVs. An outstanding example is SpliceAI [11], a 32-layer deep neural network widely recognized as the most accurate tool for predicting splicing variants currently available (e.g., [12][13][14][15]). While these in silico prediction tools are valuable, they cannot be used in isolation to establish pathogenicity in accordance with variant classification guidelines recommended by the American College of Medical Genetics and Genomics (ACMG) [16].…”

Section: Introductionmentioning

confidence: 99%

Combining full-length gene assay and SpliceAI to interpret the splicing impact of all possibleSPINK1coding variants

Wu,

Lin,

Tang

et al. 2023

Preprint

View full text Add to dashboard Cite

BackgroundSingle-nucleotide variants (SNVs) within gene coding sequences can significantly impact pre-mRNA splicing, bearing profound implications for pathogenic mechanisms and precision medicine. However, reliable splicing analysis often faces practical limitations, especially when the relevant tissues are challenging to access. Whilein silicopredictions are valuable, they alone do not meet clinical classification standards. In this study, we aim to harness the well-established full-length gene splicing assay (FLGSA) in conjunction with SpliceAI to prospectively interpret the splicing effects of all potential coding SNVs within the four-exonSPINK1gene, a gene associated with chronic pancreatitis.ResultsWe initiated the study with a retrospective correlation analysis (involving 27 previously FLGSA-analyzedSPINK1coding SNVs), progressed to a prospective correlation analysis (incorporating 35 newly FLGSA-testedSPINK1coding SNVs), followed by data extrapolation, and ended with further validation. In total, we analyzed 67SPINK1coding SNVs, representing 9.3% of all 720 possible coding SNVs and affecting 19.2% of the 240 coding nucleotides. Among these 67 FLGSA-analyzed SNVs, 12 were found to impact splicing. Through extensive cross-correlation of the FLGSA-obtained and SpliceAI-predicted data, we reasonably extrapolated that none of the unanalyzed 653 coding SNVs in theSPINK1gene are likely to exert a significant effect on splicing. Out of these 12 splice-altering events, nine produced both wild-type and aberrant transcripts, while the remaining three exclusively generated aberrant transcripts. These splice-altering SNVs were predominantly concentrated in exons 1 and 2, particularly affecting the first and/or last coding nucleotide of each exon. Among the 12 splice-altering events, 11 were missense variants, constituting 2.17% of the 506 potential missense variants, while one was synonymous, accounting for 0.61% of the 164 potential synonymous variants.ConclusionsIntegrating FLGSA with SpliceAI, we conclude that less than 2% (1.67%) of all possibleSPINK1coding SNVs have a discernible influence on splicing outcomes. Our findings underscore the importance of performing splicing analysis in the broader genomic sequence context of the study gene, highlight the inherent uncertainties associated with intermediate SpliceAI scores (i.e., those ranging from 0.20 to 0.80), and have general implications for the shift from “retrospective” to “prospective” analysis in terms of variant classification.

show abstract