Probing the Virtual Proteome to Identify Novel Disease Biomarkers

Abstract

Background: Proteomic approaches allow measurement of thousands of proteins in a single specimen, which can accelerate biomarker discovery. However, applying these technologies to massive biobanks is not currently feasible because of the practical barriers and costs of implementing such assays at scale. To overcome these challenges, we used a “virtual proteomic” approach, linking genetically predicted protein levels to clinical diagnoses in >40 000 individuals. Methods: We used genome-wide association data from the Framingham Heart Study (n=759) to construct genetic predictors for 1129 plasma protein levels. We validated the genetic predictors for 268 proteins and used them to compute predicted protein levels in 41 288 genotyped individuals in the Electronic Medical Records and Genomics (eMERGE) cohort. We tested associations for each predicted protein with 1128 clinical phenotypes. Lead associations were validated with directly measured protein levels and either low-density lipoprotein cholesterol or subclinical atherosclerosis in the MDCS (Malmö Diet and Cancer Study; n=651). Results: In the virtual proteomic analysis in eMERGE, 55 proteins were associated with 89 distinct diagnoses at a false discovery rate q<0.1. Among these, 13 associations involved lipid (n=7) or atherosclerosis (n=6) phenotypes. We tested each association for validation in MDCS using directly measured protein levels. At Bonferroni-adjusted significance thresholds, levels of apolipoprotein E isoforms were associated with hyperlipidemia, and circulating C-type lectin domain family 1 member B and platelet-derived growth factor receptor-β predicted subclinical atherosclerosis. Odds ratios for carotid atherosclerosis were 1.31 (95% CI, 1.08–1.58; P =0.006) per 1-SD increment in C-type lectin domain family 1 member B and 0.79 (0.66–0.94; P =0.008) per 1-SD increment in platelet-derived growth factor receptor-β. Conclusions: We demonstrate a biomarker discovery paradigm to identify candidate biomarkers of cardiovascular and other diseases.

Linking protein to phenotype with Mendelian Randomization detects 38 proteins with causal roles in human diseases and traits

“…It offers a data-driven approach to drug discovery using population-level data, and quantifies the strength of evidence for causation. Previous studies have made successful forays into the use of pQTL in mapping protein variation onto disease [12][13][14][15][16][17][18], and both the coverage of the proteome and the availability of disease and trait GWA study results are ever increasing. By using the lead variants of locally-acting pQTLs as instrumental variables, we focused specifically on a subset of functionally relevant variants for those proteins under study: this choice PLOS GENETICS…”

“…Previous studies have also leveraged the increased availability of pQTL data for drug target and biomarker discovery [12][13][14][15][16][17][18]. For example, in one of the largest pQTL studies to date, Sun et al [14] applied an aptamer-based approach (rather than an antibody-based assay as here) to perform extensive co-localization analyses and used MR to assess the causal contribution of IL1RL1-IL18R1 locus to atopic dermatitis, and that of MMP12 to coronary heart disease.…”

Pattern Recognition Receptor based Prognostic Biomarkers for predicting Survival of Uterine Corpus Endometrial Cancer Patients

“…Amongst these, the performance of gene voting based model based on nine genes was found to be the best. Amongst these nine genes, CLEC1B known as C-type lectin domain family 1member B found very much useful in atherosclerosis disease [57]. Aggretin AACT is a potential candidate for the treatment of tumour metastasis through CLEC-2 blockade [58].…”

Personizing the prediction of future susceptibility to a specific disease

“…Therefore, compiling biomarker data from multiple literature associated with biomarker-disease associations has become a necessity for maximize the pool of biomarkers. However, compiling such large-scale pool of biomarkers is infeasible due to the practical challenges and resource costs involved [4]. This has led some current computational methods to take advantage of the exponential increase in biomedical literature as a rich source of biomarker information [5].…”