Pathogenic scores are computed for each of the genes considered to be affected by the SV, i.e. candidate genes. Candidate genes are those genes whose regulatory domain (TAD) or sequence (e.g gene deletion) is altered by the SV. A candidate gene is not necessarily involved in the disease etiology (i.e. candidate genes include both disease causative and non-causative genes)

For each candidate gene and cell type/tissue where the SV pathogenicity is evaluated, two pathogenic scores are computed:

• Loss of Function (LOF) pathogenic score, to evaluate a pathogenic loss of gene function or expression.
• Gain of Function (GOF) pathogenic score, to evaluate a pathogenic gain of gene function or expression.

The pathogenic score, which ranges from 0 to 1, corresponds with the average of different subscores (each of them also ranging from 0 to 1):

• genePhenoScore
  Values of this score close to 1 indicate that the candidate gene has been previously associated with the SV-associated phenotype in other individuals (default).
  When running POSTRE with the advanced feature option: Gene-PatientPheno: set to No, the genePhenoScore will be 1 if the candidate gene has been associated previously with any kind of disease (regardless of whether it matches the phenotype associated with the SV).
  
  
• geneEnhancerScore
  Values of this score close to 1 indicate considerable changes in the regulatory landscape of the candidate gene. For LOF it involves the loss of cognate enhancer activity (e.g. through the deletion of some of them), and for GOF the gain of enhancer activity (either through the duplication of cognate enhancers or the adoption of ectopic enhancers). The geneEnhancerScore is only considered when the candidate gene sequence has not been directly affected by the SV.
  
  
• geneFeaturesScore
This score averages the information of two additional subscores:
• dosageSensitivityScore
  Values of this score close to 1 indicate that for the candidate gene, deviations from the normal dosage (i.e. number of copies), or expression levels, are likely detrimental. 0 means that these deviations are not detrimental.
  
  
• polycombScore
  Values of this score close to 1 indicate strong evidence for being a H3K27me3-polycomb based developmental gene (genes covered by broad domains of H3K27me3 when inactive and with tissue specific expression patterns), 0 implies the opposite. These genes tend to regulate cellular identity and altering their expression can lead to strong phenotypic alterations (Rehimi et al., 2016; Shim et al., 2020). In addition, this type of gene shows high enhancer responsiveness (Kraft et al.,2019; Pachano et al., 2021; Xu et al., 2022), which might be attributed to the presence of promoter CpG islands that prevent DNA methylation, facilitate enhancer-gene communication and, overall, provide a permissive chromatin environment.
  
  
For the geneFeaturesScore to be > 0 either the polycombScore or the dosageSensitiveScore must be >= 0.85 in the Standard running mode (default), while both scores must be >= 0.85 in the High-Specificity running mode.


• geneExpressionScore
  Values of this score close to 1 indicate a relevant gene expression status for the candidate gene. Considering the expression status of the candidate genes is particularly important for LOF situations, as the disease-causative genes must be expressed in the relevant cell types/tissues.
  
  

The pathogenic score for LOF and GOF is obtained as a result of averaging the different subscores. If any of them (pathogenic score for LOF or GOF) is higher than 0.8 it will be considered as potentially pathogenic and additional information will be provided through the results overview heatmap.

NOTE: When checking the pathogenic scores results table, some of the different subscores may have a value of "Ignored". This is because that subscore has been ignored when computing the pathogenicity. For instance, when modeling the pathogenic effect of a gene deletion and computing the pathogenic score for LOF, the geneEnhancerScore will appear as "Ignored", since long-range effects are not considered when the candidate gene sequence has been disrupted. At the same time, regarding the pathogenic score for GOF for the same scenario (gene deletion), all the subscores will be "Ignored", given that a pathogenic gain of expression is not feasible if the candidate gene has been deleted. If you want to know more about POSTRE pathogenic score calculation, please check our manuscript .