About the PIPE

  ABOUT

      Cross-disease genomic analysis has identified shared genetic loci, mostly at the non-coding genome, each exerting pleiotropic effects on two or more diseases, for example, neuropsychiatric disorders and inflammatory disorders. In the context of disease genetics, the wording pleiotropic loci refers to shared genetic loci, each exerting effects on two or more diseases. This is different from the concept of polygenic loci used to describe multiple genetic loci that are converged to affect a single disease.
      Phenotypically divergent diseases may share pathogenic mechanisms, displaying a high degree of therapeutic sharing. Identification of shared pleiotropic loci and effector genes (and pathways they function in), thus, have important implications in therapeutic target discovery and drug repurposing. The intrinsic difficulty in linking pleiotropic non-coding loci to candidate targets, however, poses a challenge (simplified as W-H-W): it is not addressed whether pleiotropic loci are useful for therapeutic target discovery; if useful, neither is it addressed how to implement pleiotropy-driven target prioritisation and evaluation; and in which diseases pleiotropy can inform clinical therapeutics.
      As a pleiotropy-driven unsupervised framework with two components (prioritisation and evaluation), Pleiotropy Informing Prioritisation & Evaluation (or PIPE) resolves pleiotropic associations for translational use in therapeutic target identification and validation. The validity (and utility) is demonstrated by showcasing two well-studied models for studying cross-disease genetic pleiotropy (because of high heritability and therapeutic sharing): neuropsychiatric disorders and inflammatory disorders.

  GLOSSARY

      cGene: conformation genes with evidence of gene promoters physically interacting with SNP-harbouring genomic regions, using promoter capture Hi-C datasets.
      Clinical proof-of-concept targets: defined as therapeutic genes that have been targeted by drugs reaching development phase 2 or above for a specific disease.
      Credit score: calculated by combining genomic predictors, ranging from 0 to 10 used to rank target genes prioritised across the genome.
      eGene: expression-associated genes with evidence of genetic regulation of gene expression, using eQTL mapping datasets.
      LPA: leading prioritisation analysis performed to quantify the degree to which a target set (e.g. clinical proof-of-concept targets) is enriched at the 'leading edge/prioritisation' of the prioritised gene list ranked by credit scores.
      nGene: nearby genes with evidence of genomic proximity.
      Pathway crosstalk: a network of highly prioritised/credited genes that interconnect to each other.
      PICT: Pleiotropy Informing Clinical Therapeutics.
      PIPE: Pleiotropy Informing Prioritisation and Evaluation.
      Simulated negative targets: simulated in a manner that, conditioned on the existing druggable landscape (i.e. evidence already explored), they are unlikely to be target genes for a specific disease.
      W-H-W: the challenges to solve the intrinsic difficulty in linking pleiotropic non-coding loci to candidate genes, that is, whether pleiotropic loci are useful for therapeutic target discovery; if useful, how to implement pleiotropy-driven target prioritisation and evaluation; and in what diseases pleiotropy can inform clinical therapeutics.