Genetic drug target validation using Mendelian randomization

Abstract

Mendelian randomisation analysis has emerged as an important tool to elucidate the causal relevance of a range of environmental and biological risk factors for human disease. However, inference on cause is undermined if the genetic variants used to instrument a risk factor of interest also associate with other traits that open alternative pathways to the disease (horizontal pleiotropy). We show how the 'no horizontal pleiotropy assumption' in MR analysis is strengthened when proteins are the risk factors of interest. Proteins are the proximal effectors of biological processes encoded in the genome, and are becoming assayable on an -omics scale.Moreover, proteins are the targets of most medicines, so Mendelian randomization (MR) studies of drug targets are becoming a fundamental tool in drug development. To enable such studies we introduce a formal mathematical framework that contrasts MR analysis of proteins with that of risk factors located more distally in the causal chain from gene to disease. Finally, we illustrate key model decisions and introduce an analytical framework for maximizing power and elucidating the robustness of drug target MR analyses.

Causal effects of circulating cytokine concentrations on risk of Alzheimer’s disease: A bidirectional two-sample Mendelian randomization study

“…F statistic Genetic variant strongly associated with exposure of interest (24) MR-Egger Adjusting for horizontal pleiotropic effects of genetic variants (25) Weighted median 50% of the genetic instruments are invalid (26) Leave-one variant out Pleiotropic effects of variants (27) Cohran's Q Heterogeneity of genetic effectspleiotropic effects (25) Steiger filtering Hypothesized causal direction for each SNP (28) Cis Mendelian randomization Pleiotropic effects of variants (29) . CC-BY 4.0 International license It is made available under a perpetuity.…”

Genomic Medicine: Lessons Learned From Monogenic and Complex Bone Disorders

“…The recent focus has been made on proteins, which are the target of most drugs, i.e., using proteomics information to leverage drug identification. Schmidt et al ( 173 ) have developed a novel MR framework for drug validation so called “cis-MR” directed at increasing the precision and robustness of the MR approach. Traditionally, MR uses genetic variants as instrumental variables that are associated with the outcome, independent from other genetic variants in the locus or located elsewhere throughout the genome.…”

“…Traditionally, MR uses genetic variants as instrumental variables that are associated with the outcome, independent from other genetic variants in the locus or located elsewhere throughout the genome. In contrast, the cis-MR approach is more stringent, only employing genetic variants located in or in the vicinity of a protein coding genes ( 173 ). Recently, work by Zheng et al, have highlighted the important role of cis and trans protein quantitative trait loci (pQTLs) MR analysis, which coupled with evidence for colocalization produces robust evidence of causal protein-phenotype associations as well ( 174 ).…”

Validation of lipid-related therapeutic targets for coronary heart disease prevention using human genetics

“…Genetic effects (like drug action) are mediated through proteins (according to Crick's Central Dogma), and variation in the genome is inherited at random from parents to offspring (according to Mendel's Laws), much like treatment allocation in a clinical trial (11). We and others have shown that variants in a gene encoding a specific drug target, that alter its expression or function, can be used as a tool to anticipate the effect of drug action on the same target (12). We have referred to this application of Mendelian randomization as 'drug target MR' (12).…”