Prioritization model to identify SRE-disrupting variants for
splicing analysis
While the current SRE-dedicated bioinformatic tools have significant
limitations in terms of specificity, our results indicate that using a
combination of tools can improve prediction and increase confidence in
the selection of variants for confirmatory splicing assays. Drawing from
the variant selection process (Figure 2), results relating to donor and
acceptor splice site strength (Table 3), and the evaluation of SRE
prediction tools (Table 4), we developed a generic workflow. This
prioritization model uses MES for native and de novo splice site
analysis, and HSF and ΔHZEI in series for SRE
prediction, as summarized in Figure 4.