Multi-response Mendelian randomization: Identification of shared and distinct exposures for multimorbidity and multiple related disease outcomes.

The existing framework of Mendelian randomization (MR) infers the causal effect of one or multiple exposures on one single outcome. It is not designed to jointly model multiple outcomes, as would be necessary to detect causes of more than one outcome and would be relevant to model multimorbidity or other related disease outcomes. Here, we introduce multi-response Mendelian randomization (MR2), an MR method specifically designed for multiple outcomes to identify exposures that cause more than one outcome or, conversely, exposures that exert their effect on distinct responses. MR2 uses a sparse Bayesian Gaussian copula regression framework to detect causal effects while estimating the residual correlation between summary-level outcomes, i.e., the correlation that cannot be explained by the exposures, and vice versa. We show both theoretically and in a comprehensive simulation study how unmeasured shared pleiotropy induces residual correlation between outcomes irrespective of sample overlap. We also reveal how non-genetic factors that affect more than one outcome contribute to their correlation. We demonstrate that by accounting for residual correlation, MR2 has higher power to detect shared exposures causing more than one outcome. It also provides more accurate causal effect estimates than existing methods that ignore the dependence between related responses. Finally, we illustrate how MR2 detects shared and distinct causal exposures for five cardiovascular diseases in two applications considering cardiometabolic and lipidomic exposures and uncovers residual correlation between summary-level outcomes reflecting known relationships between cardiovascular diseases.

Using genetic association data to guide drug discovery and development: Review of methods and applications.

Evidence on the validity of drug targets from randomized trials is reliable but typically expensive and slow to obtain. In contrast, evidence from conventional observational epidemiological studies is less reliable because of the potential for bias from confounding and reverse causation. Mendelian randomization is a quasi-experimental approach analogous to a randomized trial that exploits naturally occurring randomization in the transmission of genetic variants. In Mendelian randomization, genetic variants that can be regarded as proxies for an intervention on the proposed drug target are leveraged as instrumental variables to investigate potential effects on biomarkers and disease outcomes in large-scale observational datasets. This approach can be implemented rapidly for a range of drug targets to provide evidence on their effects and thus inform on their priority for further investigation. In this review, we present statistical methods and their applications to showcase the diverse opportunities for applying Mendelian randomization in guiding clinical development efforts, thus enabling interventions to target the right mechanism in the right population group at the right time. These methods can inform investigators on the mechanisms underlying drug effects, their related biomarkers, implications for the timing of interventions, and the population subgroups that stand to gain the most benefit. Most methods can be implemented with publicly available data on summarized genetic associations with traits and diseases, meaning that the only major limitations to their usage are the availability of appropriately powered studies for the exposure and outcome and the existence of a suitable genetic proxy for the proposed intervention.

Combining evidence from Mendelian randomization and colocalization: Review and comparison of approaches.

Mendelian randomization and colocalization are two statistical approaches that can be applied to summarized data from genome-wide association studies (GWASs) to understand relationships between traits and diseases. However, despite similarities in scope, they are different in their objectives, implementation, and interpretation, in part because they were developed to serve different scientific communities. Mendelian randomization assesses whether genetic predictors of an exposure are associated with the outcome and interprets an association as evidence that the exposure has a causal effect on the outcome, whereas colocalization assesses whether two traits are affected by the same or distinct causal variants. When considering genetic variants in a single genetic region, both approaches can be performed. While a positive colocalization finding typically implies a non-zero Mendelian randomization estimate, the reverse is not generally true: there are several scenarios which would lead to a non-zero Mendelian randomization estimate but lack evidence for colocalization. These include the existence of distinct but correlated causal variants for the exposure and outcome, which would violate the Mendelian randomization assumptions, and a lack of strong associations with the outcome. As colocalization was developed in the GWAS tradition, typically evidence for colocalization is concluded only when there is strong evidence for associations with both traits. In contrast, a non-zero estimate from Mendelian randomization can be obtained despite only nominally significant genetic associations with the outcome at the locus. In this review, we discuss how the two approaches can provide complementary information on potential therapeutic targets.

Beyond factor H: The impact of genetic-risk variants for age-related macular degeneration on circulating factor-H-like 1 and factor-H-related protein concentrations.

Age-related macular degeneration (AMD) is a leading cause of vision loss; there is strong genetic susceptibility at the complement factor H (CFH) locus. This locus encodes a series of complement regulators: factor H (FH), a splice variant factor-H-like 1 (FHL-1), and five factor-H-related proteins (FHR-1 to FHR-5), all involved in the regulation of complement factor C3b turnover. Little is known about how AMD-associated variants at this locus might influence FHL-1 and FHR protein concentrations. We have used a bespoke targeted mass-spectrometry assay to measure the circulating concentrations of all seven complement regulators and demonstrated elevated concentrations in 352 advanced AMD-affected individuals for all FHR proteins (FHR-1, p = 2.4 × 10-10; FHR-2, p = 6.0 × 10-10; FHR-3, p = 1.5 × 10-5; FHR-4, p = 1.3 × 10-3; FHR-5, p = 1.9 × 10-4) and FHL-1 (p = 4.9 × 10-4) when these individuals were compared to 252 controls, whereas no difference was seen for FH (p = 0.94). Genome-wide association analyses in controls revealed genome-wide-significant signals at the CFH locus for all five FHR proteins, and univariate Mendelian-randomization analyses strongly supported the association of FHR-1, FHR-2, FHR-4, and FHR-5 with AMD susceptibility. These findings provide a strong biochemical explanation for how genetically driven alterations in circulating FHR proteins could be major drivers of AMD and highlight the need for research into FHR protein modulation as a viable therapeutic avenue for AMD.

Risk factors for human papillomavirus infection, cervical intraepithelial neoplasia and cervical cancer: an umbrella review and follow-up Mendelian randomisation studies.

BACKGROUND: Persistent infection by oncogenic human papillomavirus (HPV) is necessary although not sufficient for development of cervical cancer. Behavioural, environmental, or comorbid exposures may promote or protect against malignant transformation. Randomised evidence is limited and the validity of observational studies describing these associations remains unclear. METHODS: In this umbrella review, we searched electronic databases to identify meta-analyses of observational studies that evaluated risk or protective factors and the incidence of HPV infection, cervical intra-epithelial neoplasia (CIN), cervical cancer incidence and mortality. Following re-analysis, evidence was classified and graded based on a pre-defined set of statistical criteria. Quality was assessed with AMSTAR-2. For all associations graded as weak evidence or above, with available genetic instruments, we also performed Mendelian randomisation to examine the potential causal effect of modifiable exposures with risk of cervical cancer. The protocol for this study was registered on PROSPERO (CRD42020189995). RESULTS: We included 171 meta-analyses of different exposure contrasts from 50 studies. Systemic immunosuppression including HIV infection (RR = 2.20 (95% CI = 1.89-2.54)) and immunosuppressive medications for inflammatory bowel disease (RR = 1.33 (95% CI = 1.27-1.39)), as well as an altered vaginal microbiome (RR = 1.59 (95% CI = 1.40-1.81)), were supported by strong and highly suggestive evidence for an association with HPV persistence, CIN or cervical cancer. Smoking, number of sexual partners and young age at first pregnancy were supported by highly suggestive evidence and confirmed by Mendelian randomisation. CONCLUSIONS: Our main analysis supported the association of systemic (HIV infection, immunosuppressive medications) and local immunosuppression (altered vaginal microbiota) with increased risk for worse HPV and cervical disease outcomes. Mendelian randomisation confirmed the link for genetically predicted lifetime smoking index, and young age at first pregnancy with cervical cancer, highlighting also that observational evidence can hide different inherent biases. This evidence strengthens the need for more frequent HPV screening in people with immunosuppression, further investigation of the vaginal microbiome and access to sexual health services.

Prioritizing the Role of Major Lipoproteins and Subfractions as Risk Factors for Peripheral Artery Disease.

BACKGROUND: Lipoprotein-related traits have been consistently identified as risk factors for atherosclerotic cardiovascular disease, largely on the basis of studies of coronary artery disease (CAD). The relative contributions of specific lipoproteins to the risk of peripheral artery disease (PAD) have not been well defined. We leveraged large-scale genetic association data to investigate the effects of circulating lipoprotein-related traits on PAD risk. METHODS: Genome-wide association study summary statistics for circulating lipoprotein-related traits were used in the mendelian randomization bayesian model averaging framework to prioritize the most likely causal major lipoprotein and subfraction risk factors for PAD and CAD. Mendelian randomization was used to estimate the effect of apolipoprotein B (ApoB) lowering on PAD risk using gene regions proxying lipid-lowering drug targets. Genes relevant to prioritized lipoprotein subfractions were identified with transcriptome-wide association studies. RESULTS: ApoB was identified as the most likely causal lipoprotein-related risk factor for both PAD (marginal inclusion probability, 0.86; P=0.003) and CAD (marginal inclusion probability, 0.92; P=0.005). Genetic proxies for ApoB-lowering medications were associated with reduced risk of both PAD (odds ratio,0.87 per 1-SD decrease in ApoB [95% CI, 0.84-0.91]; P=9×10-10) and CAD (odds ratio,0.66 [95% CI, 0.63-0.69]; P=4×10-73), with a stronger predicted effect of ApoB lowering on CAD (ratio of effects, 3.09 [95% CI, 2.29-4.60]; P<1×10-6). Extra-small very-low-density lipoprotein particle concentration was identified as the most likely subfraction associated with PAD risk (marginal inclusion probability, 0.91; P=2.3×10-4), whereas large low-density lipoprotein particle concentration was the most likely subfraction associated with CAD risk (marginal inclusion probability, 0.95; P=0.011). Genes associated with extra-small very-low-density lipoprotein particle and large low-density lipoprotein particle concentration included canonical ApoB pathway components, although gene-specific effects were variable. Lipoprotein(a) was associated with increased risk of PAD independently of ApoB (odds ratio, 1.04 [95% CI, 1.03-1.04]; P=1.0×10-33). CONCLUSIONS: ApoB was prioritized as the major lipoprotein fraction causally responsible for both PAD and CAD risk. However, ApoB-lowering drug targets and ApoB-containing lipoprotein subfractions had diverse associations with atherosclerotic cardiovascular disease, and distinct subfraction-associated genes suggest possible differences in the role of lipoproteins in the pathogenesis of PAD and CAD.

The relationship between lipoprotein A and other lipids with prostate cancer risk: A multivariable Mendelian randomisation study.

BACKGROUND: Numerous epidemiological studies have investigated the role of blood lipids in prostate cancer (PCa) risk, though findings remain inconclusive to date. The ongoing research has mainly involved observational studies, which are often prone to confounding. This study aimed to identify the relationship between genetically predicted blood lipid concentrations and PCa. METHODS AND FINDINGS: Data for low-density lipoprotein (LDL) cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides (TG), apolipoprotein A (apoA) and B (apoB), lipoprotein A (Lp(a)), and PCa were acquired from genome-wide association studies in UK Biobank and the PRACTICAL consortium, respectively. We used a two-sample summary-level Mendelian randomisation (MR) approach with both univariable and multivariable (MVMR) models and utilised a variety of robust methods and sensitivity analyses to assess the possibility of MR assumptions violation. No association was observed between genetically predicted concentrations of HDL, TG, apoA and apoB, and PCa risk. Genetically predicted LDL concentration was positively associated with total PCa in the univariable analysis, but adjustment for HDL, TG, and Lp(a) led to a null association. Genetically predicted concentration of Lp(a) was associated with higher total PCa risk in the univariable (ORweighted median per standard deviation (SD) = 1.091; 95% CI 1.028 to 1.157; P = 0.004) and MVMR analyses after adjustment for the other lipid traits (ORIVW per SD = 1.068; 95% CI 1.005 to 1.134; P = 0.034). Genetically predicted Lp(a) was also associated with advanced (MVMR ORIVW per SD = 1.078; 95% CI 0.999 to 1.163; P = 0.055) and early age onset PCa (MVMR ORIVW per SD = 1.150; 95% CI 1.015,1.303; P = 0.028). Although multiple estimation methods were utilised to minimise the effect of pleiotropy, the presence of any unmeasured pleiotropy cannot be excluded and may limit our findings. CONCLUSIONS: We observed that genetically predicted Lp(a) concentrations were associated with an increased PCa risk. Future studies are required to understand the underlying biological pathways of this finding, as it may inform PCa prevention through Lp(a)-lowering strategies.

Associations of genetically predicted fatty acid levels across the phenome: A mendelian randomisation study.

BACKGROUND: Fatty acids are important dietary factors that have been extensively studied for their implication in health and disease. Evidence from epidemiological studies and randomised controlled trials on their role in cardiovascular, inflammatory, and other diseases remains inconsistent. The objective of this study was to assess whether genetically predicted fatty acid concentrations affect the risk of disease across a wide variety of clinical health outcomes. METHODS AND FINDINGS: The UK Biobank (UKB) is a large study involving over 500,000 participants aged 40 to 69 years at recruitment from 2006 to 2010. We used summary-level data for 117,143 UKB samples (base dataset), to extract genetic associations of fatty acids, and individual-level data for 322,232 UKB participants (target dataset) to conduct our discovery analysis. We studied potentially causal relationships of circulating fatty acids with 845 clinical diagnoses, using mendelian randomisation (MR) approach, within a phenome-wide association study (PheWAS) framework. Regression models in PheWAS were adjusted for sex, age, and the first 10 genetic principal components. External summary statistics were used for replication. When several fatty acids were associated with a health outcome, multivariable MR and MR-Bayesian method averaging (MR-BMA) was applied to disentangle their causal role. Genetic predisposition to higher docosahexaenoic acid (DHA) was associated with cholelithiasis and cholecystitis (odds ratio per mmol/L: 0.76, 95% confidence interval: 0.66 to 0.87). This was supported in replication analysis (FinnGen study) and by the genetically predicted omega-3 fatty acids analyses. Genetically predicted linoleic acid (LA), omega-6, polyunsaturated fatty acids (PUFAs), and total fatty acids (total FAs) showed positive associations with cardiovascular outcomes with support from replication analysis. Finally, higher genetically predicted levels of DHA (0.83, 0.73 to 0.95) and omega-3 (0.83, 0.75 to 0.92) were found to have a protective effect on obesity, which was supported using body mass index (BMI) in the GIANT consortium as replication analysis. Multivariable MR analysis suggested a direct detrimental effect of LA (1.64, 1.07 to 2.50) and omega-6 fatty acids (1.81, 1.06 to 3.09) on coronary heart disease (CHD). MR-BMA prioritised LA and omega-6 fatty acids as the top risk factors for CHD. Although we present a range of sensitivity analyses to the address MR assumptions, horizontal pleiotropy may still bias the reported associations and further evaluation in clinical trials is needed. CONCLUSIONS: Our study suggests potentially protective effects of circulating DHA and omega-3 concentrations on cholelithiasis and cholecystitis and on obesity, highlighting the need to further assess them as prevention treatments in clinical trials. Moreover, our findings do not support the supplementation of unsaturated fatty acids for cardiovascular disease prevention.

Identifying and ranking causal biochemical biomarkers for breast cancer: a Mendelian randomisation study.

BACKGROUND: Only a few of the 34 biochemical biomarkers measured in the UK Biobank (UKB) have been associated with breast cancer, with many associations suffering from possible confounding and reverse causation. This study aimed to screen and rank all UKB biochemical biomarkers for possible causal relationships with breast cancer. METHODS: We conducted two-sample Mendelian randomisation (MR) analyses on ~420,000 women by leveraging summary-level genetic exposure associations from the UKB study (n = 194,174) and summary-level genetic outcome associations from the Breast Cancer Association Consortium (n = 228,951). Our exposures included all 34 biochemical biomarkers in the UKB, and our outcomes were overall, oestrogen-positive, and oestrogen-negative breast cancer. We performed inverse-variance weighted MR, weighted median MR, MR-Egger, and MR-PRESSO for 30 biomarkers for which we found multiple instrumental variables. We additionally performed multivariable MR to adjust for known risk factors, bidirectional MR to investigate reverse causation, and MR Bayesian model averaging to rank the significant biomarkers by their genetic evidence. RESULTS: Increased genetic liability to overall breast cancer was robustly associated with the following biomarkers by decreasing importance: testosterone (odds ratio (OR): 1.12, 95% confidence interval (CI): 1.04-1.21), high-density lipoprotein (HDL) cholesterol (OR: 1.08, 95% CI: 1.04-1.13), insulin-like growth factor 1 (OR: 1.08, 95% CI: 1.02-1.13), and alkaline phosphatase (ALP) (OR: 0.93, 95% CI: 0.89-0.98). CONCLUSIONS: Our findings support a likely causal role of genetically predicted levels of testosterone, HDL cholesterol, and IGF-1, as well as a novel potential role of ALP in breast cancer aetiology. Further studies are needed to understand full disease pathways that may inform breast cancer prevention.

Exploring the causal effect of maternal pregnancy adiposity on offspring adiposity: Mendelian randomisation using polygenic risk scores.

BACKGROUND: Greater maternal adiposity before or during pregnancy is associated with greater offspring adiposity throughout childhood, but the extent to which this is due to causal intrauterine or periconceptional mechanisms remains unclear. Here, we use Mendelian randomisation (MR) with polygenic risk scores (PRS) to investigate whether associations between maternal pre-/early pregnancy body mass index (BMI) and offspring adiposity from birth to adolescence are causal. METHODS: We undertook confounder adjusted multivariable (MV) regression and MR using mother-offspring pairs from two UK cohorts: Avon Longitudinal Study of Parents and Children (ALSPAC) and Born in Bradford (BiB). In ALSPAC and BiB, the outcomes were birthweight (BW; N = 9339) and BMI at age 1 and 4 years (N = 8659 to 7575). In ALSPAC only we investigated BMI at 10 and 15 years (N = 4476 to 4112) and dual-energy X-ray absorptiometry (DXA) determined fat mass index (FMI) from age 10-18 years (N = 2659 to 3855). We compared MR results from several PRS, calculated from maternal non-transmitted alleles at between 29 and 80,939 single nucleotide polymorphisms (SNPs). RESULTS: MV and MR consistently showed a positive association between maternal BMI and BW, supporting a moderate causal effect. For adiposity at most older ages, although MV estimates indicated a strong positive association, MR estimates did not support a causal effect. For the PRS with few SNPs, MR estimates were statistically consistent with the null, but had wide confidence intervals so were often also statistically consistent with the MV estimates. In contrast, the largest PRS yielded MR estimates with narrower confidence intervals, providing strong evidence that the true causal effect on adolescent adiposity is smaller than the MV estimates (Pdifference = 0.001 for 15-year BMI). This suggests that the MV estimates are affected by residual confounding, therefore do not provide an accurate indication of the causal effect size. CONCLUSIONS: Our results suggest that higher maternal pre-/early-pregnancy BMI is not a key driver of higher adiposity in the next generation. Thus, they support interventions that target the whole population for reducing overweight and obesity, rather than a specific focus on women of reproductive age.

Circulating inflammatory cytokines and risk of five cancers: a Mendelian randomization analysis.

BACKGROUND: Epidemiological and experimental evidence has linked chronic inflammation to cancer aetiology. It is unclear whether associations for specific inflammatory biomarkers are causal or due to bias. In order to examine whether altered genetically predicted concentration of circulating cytokines are associated with cancer development, we performed a two-sample Mendelian randomisation (MR) analysis. METHODS: Up to 31,112 individuals of European descent were included in genome-wide association study (GWAS) meta-analyses of 47 circulating cytokines. Single nucleotide polymorphisms (SNPs) robustly associated with the cytokines, located in or close to their coding gene (cis), were used as instrumental variables. Inverse-variance weighted MR was used as the primary analysis, and the MR assumptions were evaluated in sensitivity and colocalization analyses and a false discovery rate (FDR) correction for multiple comparisons was applied. Corresponding germline GWAS summary data for five cancer outcomes (breast, endometrial, lung, ovarian, and prostate), and their subtypes were selected from the largest cancer-specific GWASs available (cases ranging from 12,906 for endometrial to 133,384 for breast cancer). RESULTS: There was evidence of inverse associations of macrophage migration inhibitory factor with breast cancer (OR per SD = 0.88, 95% CI 0.83 to 0.94), interleukin-1 receptor antagonist with endometrial cancer (0.86, 0.80 to 0.93), interleukin-18 with lung cancer (0.87, 0.81 to 0.93), and beta-chemokine-RANTES with ovarian cancer (0.70, 0.57 to 0.85) and positive associations of monokine induced by gamma interferon with endometrial cancer (3.73, 1.86 to 7.47) and cutaneous T-cell attracting chemokine with lung cancer (1.51, 1.22 to 1.87). These associations were similar in sensitivity analyses and supported in colocalization analyses. CONCLUSIONS: Our study adds to current knowledge on the role of specific inflammatory biomarker pathways in cancer aetiology. Further validation is needed to assess the potential of these cytokines as pharmacological or lifestyle targets for cancer prevention.

Genetic variation in cervical preinvasive and invasive disease: a genome-wide association study.

BACKGROUND: Most uterine cervical high-risk human papillomavirus (HPV) infections are transient, with only a small fraction developing into cervical cancer. Family aggregation studies and heritability estimates suggest a significant inherited genetic component. Candidate gene studies and previous genome-wide association studies (GWASs) report associations between the HLA region and cervical cancer. Adopting a genome-wide approach, we aimed to compare genetic variation in women with invasive cervical cancer and cervical intraepithelial neoplasia (CIN) grade 3 with that in healthy controls. METHODS: We did a GWAS in a cohort of unrelated European individuals using data from UK Biobank, a population-based cohort including 273 377 women aged 40-69 years at recruitment between March 13, 2006, and Oct 1, 2010. We used an additive univariate logistic regression model to analyse genetic variants associated with invasive cervical cancer or CIN3. We sought replication of candidate associations in FinnGen, a large independent dataset of 128 123 individuals. We also did a two-sample mendelian randomisation approach to explore the role of risk factors in the genetic risk of cervical cancer. FINDINGS: We included 4769 CIN3 and invasive cervical cancer case samples and 145 545 control samples in the GWAS. Of 9 600 464 assayed and imputed single-nucleotide polymorphisms (SNPs), six independent variants were associated with CIN3 and invasive cervical cancer. These included novel loci rs10175462 (PAX8; odds ratio [OR] 0·87, 95% CI 0·84-0·91; p=1·07 × 10-9) and rs27069 (CLPTM1L; 0·88, 0·84-0·92; p=2·51 × 10-9), and previously reported signals at rs9272050 (HLA-DQA1; 1·27, 1·21-1·32; p=2·51 × 10-28), rs6938453 (MICA; 0·79, 0·75-0·83; p=1·97 × 10-17), rs55986091 (HLA-DQB1; 0·66, 0·60-0·72; p=6·42 × 10-28), and rs9266183 (HLA-B; 0·73, 0·64-0·83; p=1·53 × 10-6). Three SNPs were replicated in the independent Finnish dataset of 1648 invasive cervical cancer cases: PAX8 (rs10175462; p=0·015), CLPTM1L (rs27069; p=2·54 × 10-7), and HLA-DQA1 (rs9272050; p=7·90 × 10-8). Mendelian randomisation further supported the complementary role of smoking (OR 2·46, 95% CI 1·64-3·69), older age at first pregnancy (0·80, 0·68-0·95), and number of sexual partners (1·95, 1·44-2·63) in the risk of developing cervical cancer. INTERPRETATION: Our results provide new evidence for the genetic susceptibility to cervical cancer, specifically the PAX8, CLPTM1L, and HLA genes, suggesting disruption in apoptotic and immune function pathways. Future studies integrating host and viral, genetic, and epigenetic variation, could further elucidate complex host-viral interactions. FUNDING: NIHR Imperial BRC Wellcome 4i Clinician Scientist Training Programme.

Leveraging human genetic data to investigate the cardiometabolic effects of glucose-dependent insulinotropic polypeptide signalling.

AIMS/HYPOTHESIS: The aim of this study was to leverage human genetic data to investigate the cardiometabolic effects of glucose-dependent insulinotropic polypeptide (GIP) signalling. METHODS: Data were obtained from summary statistics of large-scale genome-wide association studies. We examined whether genetic associations for type 2 diabetes liability in the GIP and GIPR genes co-localised with genetic associations for 11 cardiometabolic outcomes. For those outcomes that showed evidence of co-localisation (posterior probability >0.8), we performed Mendelian randomisation analyses to estimate the association of genetically proxied GIP signalling with risk of cardiometabolic outcomes, and to test whether this exceeded the estimate observed when considering type 2 diabetes liability variants from other regions of the genome. RESULTS: Evidence of co-localisation with genetic associations of type 2 diabetes liability at both the GIP and GIPR genes was observed for five outcomes. Mendelian randomisation analyses provided evidence for associations of lower genetically proxied type 2 diabetes liability at the GIP and GIPR genes with lower BMI (estimate in SD units -0.16, 95% CI -0.30, -0.02), C-reactive protein (-0.13, 95% CI -0.19, -0.08) and triacylglycerol levels (-0.17, 95% CI -0.22, -0.12), and higher HDL-cholesterol levels (0.19, 95% CI 0.14, 0.25). For all of these outcomes, the estimates were greater in magnitude than those observed when considering type 2 diabetes liability variants from other regions of the genome. CONCLUSIONS/INTERPRETATION: This study provides genetic evidence to support a beneficial role of sustained GIP signalling on cardiometabolic health greater than that expected from improved glycaemic control alone. Further clinical investigation is warranted. DATA AVAILABILITY: All data used in this study are publicly available. The scripts for the analysis are available at: https://github.com/vkarhune/GeneticallyProxiedGIP .

Inferring Causal Relationships Between Risk Factors and Outcomes from Genome-Wide Association Study Data.

An observational correlation between a suspected risk factor and an outcome does not necessarily imply that interventions on levels of the risk factor will have a causal impact on the outcome (correlation is not causation). If genetic variants associated with the risk factor are also associated with the outcome, then this increases the plausibility that the risk factor is a causal determinant of the outcome. However, if the genetic variants in the analysis do not have a specific biological link to the risk factor, then causal claims can be spurious. We review the Mendelian randomization paradigm for making causal inferences using genetic variants. We consider monogenic analysis, in which genetic variants are taken from a single gene region, and polygenic analysis, which includes variants from multiple regions. We focus on answering two questions: When can Mendelian randomization be used to make reliable causal inferences, and when can it be used to make relevant causal inferences?

Risk factors mediating the effect of body mass index and waist-to-hip ratio on cardiovascular outcomes: Mendelian randomization analysis.

BACKGROUND: Higher body mass index (BMI) and waist-to-hip ratio (WHR) increase the risk of cardiovascular disease, but the extent to which this is mediated by blood pressure, diabetes, lipid traits, and smoking is not fully understood. METHODS: Using consortia and UK Biobank genetic association summary data from 140,595 to 898,130 participants predominantly of European ancestry, Mendelian randomization mediation analysis was performed to investigate the degree to which systolic blood pressure (SBP), diabetes, lipid traits, and smoking mediated an effect of BMI and WHR on the risk of coronary artery disease (CAD), peripheral artery disease (PAD) and stroke. RESULTS: The odds ratio of CAD per 1-standard deviation increase in genetically predicted BMI was 1.49 (95% CI 1.39 to 1.60). This attenuated to 1.34 (95% CI 1.24 to 1.45) after adjusting for genetically predicted SBP (proportion mediated 27%, 95% CI 3% to 50%), to 1.27 (95% CI 1.17 to 1.37) after adjusting for genetically predicted diabetes (41% mediated, 95% CI 18% to 63%), to 1.47 (95% CI 1.36 to 1.59) after adjusting for genetically predicted lipids (3% mediated, 95% -23% to 29%), and to 1.46 (95% CI 1.34 to 1.58) after adjusting for genetically predicted smoking (6% mediated, 95% CI -20% to 32%). Adjusting for all the mediators together, the estimate attenuated to 1.14 (95% CI 1.04 to 1.26; 66% mediated, 95% CI 42% to 91%). A similar pattern was observed when considering genetically predicted WHR as the exposure, and PAD or stroke as the outcome. CONCLUSIONS: Measures to reduce obesity will lower the risk of cardiovascular disease primarily by impacting downstream metabolic risk factors, particularly diabetes and hypertension. Reduction of obesity prevalence alongside control and management of its mediators is likely to be most effective for minimizing the burden of obesity.

Shared mechanisms between coronary heart disease and depression: findings from a large UK general population-based cohort.

While comorbidity between coronary heart disease (CHD) and depression is evident, it is unclear whether the two diseases have shared underlying mechanisms. We performed a range of analyses in 367,703 unrelated middle-aged participants of European ancestry from UK Biobank, a population-based cohort study, to assess whether comorbidity is primarily due to genetic or environmental factors, and to test whether cardiovascular risk factors and CHD are likely to be causally related to depression using Mendelian randomization. We showed family history of heart disease was associated with a 20% increase in depression risk (95% confidence interval [CI] 16-24%, p < 0.0001), but a genetic risk score that is strongly associated with CHD risk was not associated with depression. An increase of 1 standard deviation in the CHD genetic risk score was associated with 71% higher CHD risk, but 1% higher depression risk (95% CI 0-3%; p = 0.11). Mendelian randomization analyses suggested that triglycerides, interleukin-6 (IL-6), and C-reactive protein (CRP) are likely causal risk factors for depression. The odds ratio for depression per standard deviation increase in genetically-predicted triglycerides was 1.18 (95% CI 1.09-1.27; p = 2 × 10-5); per unit increase in genetically-predicted log-transformed IL-6 was 0.74 (95% CI 0.62-0.89; p = 0.0012); and per unit increase in genetically-predicted log-transformed CRP was 1.18 (95% CI 1.07-1.29; p = 0.0009). Our analyses suggest that comorbidity between depression and CHD arises largely from shared environmental factors. IL-6, CRP and triglycerides are likely to be causally linked with depression, so could be targets for treatment and prevention of depression.

MethylCal: Bayesian calibration of methylation levels.

Bisulfite amplicon sequencing has become the primary choice for single-base methylation quantification of multiple targets in parallel. The main limitation of this technology is a preferential amplification of an allele and strand in the PCR due to methylation state. This effect, known as 'PCR bias', causes inaccurate estimation of the methylation levels and calibration methods based on standard controls have been proposed to correct for it. Here, we present a Bayesian calibration tool, MethylCal, which can analyse jointly all CpGs within a CpG island (CGI) or a Differentially Methylated Region (DMR), avoiding 'one-at-a-time' CpG calibration. This enables more precise modeling of the methylation levels observed in the standard controls. It also provides accurate predictions of the methylation levels not considered in the controlled experiment, a feature that is paramount in the derivation of the corrected methylation degree. We tested the proposed method on eight independent assays (two CpG islands and six imprinting DMRs) and demonstrated its benefits, including the ability to detect outliers. We also evaluated MethylCal's calibration in two practical cases, a clinical diagnostic test on 18 patients potentially affected by Beckwith-Wiedemann syndrome, and 17 individuals with celiac disease. The calibration of the methylation levels obtained by MethylCal allows a clearer identification of patients undergoing loss or gain of methylation in borderline cases and could influence further clinical or treatment decisions.

Correlation-adjusted regression survival scores for high-dimensional variable selection.

BACKGROUND: The development of classification methods for personalized medicine is highly dependent on the identification of predictive genetic markers. In survival analysis, it is often necessary to discriminate between influential and noninfluential markers. It is common to perform univariate screening using Cox scores, which quantify the associations between survival and each of the markers to provide a ranking. Since Cox scores do not account for dependencies between the markers, their use is suboptimal in the presence of highly correlated markers. METHODS: As an alternative to the Cox score, we propose the correlation-adjusted regression survival (CARS) score for right-censored survival outcomes. By removing the correlations between the markers, the CARS score quantifies the associations between the outcome and the set of "decorrelated" marker values. Estimation of the scores is based on inverse probability weighting, which is applied to log-transformed event times. For high-dimensional data, estimation is based on shrinkage techniques. RESULTS: The consistency of the CARS score is proven under mild regularity conditions. In simulations with high correlations, survival models based on CARS score rankings achieved higher areas under the precision-recall curve than competing methods. Two example applications on prostate and breast cancer confirmed these results. CARS scores are implemented in the R package carSurv. CONCLUSIONS: In research applications involving high-dimensional genetic data, the use of CARS scores for marker selection is a favorable alternative to Cox scores even when correlations between covariates are low. Having a straightforward interpretation and low computational requirements, CARS scores are an easy-to-use screening tool in personalized medicine research.

Leveraging Genomic Annotations and Pleiotropic Enrichment for Improved Replication Rates in Schizophrenia GWAS.

Most of the genetic architecture of schizophrenia (SCZ) has not yet been identified. Here, we apply a novel statistical algorithm called Covariate-Modulated Mixture Modeling (CM3), which incorporates auxiliary information (heterozygosity, total linkage disequilibrium, genomic annotations, pleiotropy) for each single nucleotide polymorphism (SNP) to enable more accurate estimation of replication probabilities, conditional on the observed test statistic ("z-score") of the SNP. We use a multiple logistic regression on z-scores to combine information from auxiliary information to derive a "relative enrichment score" for each SNP. For each stratum of these relative enrichment scores, we obtain nonparametric estimates of posterior expected test statistics and replication probabilities as a function of discovery z-scores, using a resampling-based approach that repeatedly and randomly partitions meta-analysis sub-studies into training and replication samples. We fit a scale mixture of two Gaussians model to each stratum, obtaining parameter estimates that minimize the sum of squared differences of the scale-mixture model with the stratified nonparametric estimates. We apply this approach to the recent genome-wide association study (GWAS) of SCZ (n = 82,315), obtaining a good fit between the model-based and observed effect sizes and replication probabilities. We observed that SNPs with low enrichment scores replicate with a lower probability than SNPs with high enrichment scores even when both they are genome-wide significant (p < 5x10-8). There were 693 and 219 independent loci with model-based replication rates ≥80% and ≥90%, respectively. Compared to analyses not incorporating relative enrichment scores, CM3 increased out-of-sample yield for SNPs that replicate at a given rate. This demonstrates that replication probabilities can be more accurately estimated using prior enrichment information with CM3.

Mendelian randomization with fine-mapped genetic data: Choosing from large numbers of correlated instrumental variables.

Mendelian randomization uses genetic variants to make causal inferences about the effect of a risk factor on an outcome. With fine-mapped genetic data, there may be hundreds of genetic variants in a single gene region any of which could be used to assess this causal relationship. However, using too many genetic variants in the analysis can lead to spurious estimates and inflated Type 1 error rates. But if only a few genetic variants are used, then the majority of the data is ignored and estimates are highly sensitive to the particular choice of variants. We propose an approach based on summarized data only (genetic association and correlation estimates) that uses principal components analysis to form instruments. This approach has desirable theoretical properties: it takes the totality of data into account and does not suffer from numerical instabilities. It also has good properties in simulation studies: it is not particularly sensitive to varying the genetic variants included in the analysis or the genetic correlation matrix, and it does not have greatly inflated Type 1 error rates. Overall, the method gives estimates that are less precise than those from variable selection approaches (such as using a conditional analysis or pruning approach to select variants), but are more robust to seemingly arbitrary choices in the variable selection step. Methods are illustrated by an example using genetic associations with testosterone for 320 genetic variants to assess the effect of sex hormone related pathways on coronary artery disease risk, in which variable selection approaches give inconsistent inferences.