Background
The obesity pandemic
It is no secret that being overweight is increasingly becoming the norm, with the latest research briefing in the houses of parliament claiming that 28.7% of adults in the UK are obese and a further 35.6% are overweight [1]. Obesity measured as a BMI > 30 kg/m2 has been linked to a plethora of ailments through observational studies [2–6]. According to the Office for National Statistics, in 2018, the leading cause of deaths in the UK was ischaemic heart disease for men, accounting for 13.2% of deaths [7].
BMI has been implicated as a risk factor for heart dis- ease as a whole [8]. Many studies, however, only seem to tackle more generalised terms of the disease. A previous observational study has identified a link between heart attacks and BMI [4]. They observed a cohort of 899 obese individuals in adults between 35 and 74. After 10 years of observations, the study concluded that obesity was not an independent cardiovascular risk factor. This study is statistically underpowered due to small sample size. Furthermore, like other observational studies, an obvious issue is that it may suffer from problems of con-
* Correspondence: hui.guo@manchester.ac.uk
Centre for Biostatistics, Division of Population Health, Health Services Research and Primary Care, Faculty of Medicine, Biology and Health, The University of Manchester, Manchester, UK founding (e.g. smoking, alcohol abuse) [9] and other sources of bias [10]. Conclusions from these studies have
© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data.
limited use as clinical development of treatments re- quires well-targeted causal factors [11–13]. In our study, we can put this statement to the test as we employ the use of Mendelian randomisation (MR) to bypass these issues, with the aim to better infer causality of obesity on heart attacks in a population of white European individuals.
Methods
Study design
In our investigation, we have applied a two-sample MR to explore if an elevated level of BMI (regarded as expos- ure) causes increased risk of heart attacks (regarded as outcome) using BMI-associated single nucleotide polymorphisms (SNPs) as the instruments.
Two-sample MR requires participants from two separ- ate studies – one for exposure and the other for out- come, where the individuals do not overlap since overlapping data sets would lead to our results suffering from inflated type 1 error rates [14]. However, the two samples must be representative of the same population. Thus, we have used summary data of European descen- dants from two independent studies: Genetic Investiga- tion of Anthropometric Traits (GIANT) consortium (for BMI) [15] and the UK Biobank (for heart attacks) [16].
BMI data - GIANT
The genetic instruments were SNPs selected from GIANT [15]. This study consists of a meta-analysis of a population from European ancestry containing 322,154 individuals from 114 studies. Summary level data was extracted (see Summary Data section for more details) for BMI associated SNPs (P <5× 10− 8). These SNPs were further filtered by clumping carried out using the MR-Base platform [17] to ensure that the final set of instruments in MR analysis were independent of one an- other. Essentially the SNPs in linkage disequilibrium (LD) (R2 ≥ 0.001) were clumped together with only the SNP with the lowest p-value being retained.
Heart attack data - UK biobank
The UK Biobank data contains approximately half a mil- lion individuals aged between 40 and 69 years [18]. The participants were recruited from across the UK between 2006 and 2010 and asked to provide information via questionnaires, interviews, anthropometric measures and samples (e.g. blood, urine and saliva). The summary data we used for heart attacks was from GWAS results by the Neale lab who carried out rigorous quality control (QC) checks [16]. These checks whittled down the individuals involved to only QC positives (n = 337,199). The filter which caused the largest reduction in participants was the restriction to white British genetic ancestry only. Participants were also removed if they were closely related to other individuals in the study or had sex chromosome abnormalities. To learn more about the QC process please see the Neale Lab website [16].
As of 2018, over 92 million autosomal SNPs (directly genotyped or imputed) were available for analysis. All these SNPs were further restricted by minor allele fre- quency (MAF) 0.1%, Hardy-Weinberg Equilibrium (HWE) p-value > 1 × 10− 10 in the QC positive individ- uals and an imputation score INFO > 0.8 leaving ap- proximately 13.8 million SNPs for analysis [16]
For the heart attack data obtained from the UK Biobank, participants were asked in a survey on their medical history to categorically state if they had had a doctor diagnosed heart attack or stroke, or suffered from angina or high blood pressure [19]. Participants could also state if they had suffered from none of the above. The data from this survey was converted into binary (1: suffered from a doctor diagnosed heart attack, 0 otherwise).
Summary data
Instead of using individual-level data, one of the advan- tages of MR analysis is leverage summary statistics at the SNP level (estimated SNP effects, standard errors and corresponding p-values from regression models, effect alleles and other alleles along with their frequencies). These summary statistics from many large-scale genome wide association studies (GWAS) are now made publicly available.
Summary data of BMI and heart attacks were ex- tracted separately, for BMI associated SNPs, from the GIANT and UK Biobank studies [15, 16]. In GIANT, data were standardised such that per unit change in BMI corresponds to 1 standard deviation (or 4.5 kg/m2) change in BMI. We then carried out harmonisation using the TwoSample MR package in R to make sure that the effects of a SNP on the outcome and exposure were relative to the same allele, which produced one merged dataset for our MR analysis.
Statistical analysis
Before the advent of MR, observational studies were greatly limited by problems of unobserved confounding. These limitations rendered many findings lack of causal interpretations. MR, however, circumvents these difficul- ties by mimicking randomised controlled experiment and assuming that the instruments (SNPs) fulfil three criteria listed in Fig. 1.
We used two MR methods: inverse variance weighted (IVW) estimation and MR robust adjusted profile score (RAPS) to estimate causal effect of BMI on heart attacks (log odds ratio), its standard error and corresponding p- value [20, 21]. Both of the methods require that instru- mental SNPs are associated with the exposure and
CS 340 Milestone One Guidelines and Rubric Overview: For this assignment, you will implement the fundamental operations of create, read, update,
Retail Transaction Programming Project Project Requirements: Develop a program to emulate a purchase transaction at a retail store. This
7COM1028 Secure Systems Programming Referral Coursework: Secure
Create a GUI program that:Accepts the following from a user:Item NameItem QuantityItem PriceAllows the user to create a file to store the sales receip
CS 340 Final Project Guidelines and Rubric Overview The final project will encompass developing a web service using a software stack and impleme
