Fig. 1

Overview of SvAnna algorithm. A Sequence deleteriousness score δ(G). The score assesses deleteriousness (predicted effect on gene function) by means of a series of heuristics for different SV classes (Table 1). B Phenotype similarity score Φ(Q,D). SvAnna calculates the phenotypic similarity for a set of HPO terms Q representing the patient’s phenotypic features and HPO terms D for a disease. SvAnna computes the information content (IC) of the most informative common ancestor (MICA) for all term pairs q, d for q ∈ Q and d ∈ D. The mean ICs μ_Q and μ_D are calculated for Q and D, and the final similarity score Φ is calculated as the mean of μ_D and μ_Q. The δ(G) and the Φ(Q,D) scores are combined to obtain the final PSV score (Methods)