MR Dictionary

Non-overlapping samples (in two-sample MR)

The strengths of two-sample MR in terms of weak instrument bias being towards the null and minimising overfitting of the data assume that the two samples are completely independent of each other (i.e., there is no overlap of participants between the two samples). 

In practice, when summary-level data from publicly available data are used, it may not be possible to determine whether there is overlap between samples. This is because many genome-wide association studies (GWASs) are conducted on consortia of many studies and, increasingly, GWASs are beginning to consistently utilized large-scale biobanks (e.g., UK Biobank). Thus, there is potential for overlap between some of the studies (and participants) that contribute to these consortia (such as the GIANT consortia GWAS for body mass index, waist-hip ratio and height, and the DIAGRAM consortia GWAS for diabetes). The likelihood that any weak instrument bias within a two-sample setting will resemble the one-sample setting, whereby the MR estimates will be biased towards the (possibly confounded) observational estimate, is approximately proportional to the amount of overlap. Therefore, if the genetic instrumental variables (IVs) are strong, participant overlap is less of a concern. Careful reading of consortia websites and supplementary material should be undertaken to determine which studies contribute to each of the samples used to characterise the associations between the IV-exposure and IV-outcome and, if possible, sensitivity analyses undertaken with overlapping studies removed. Additionally, methods such as bivariate linkage disequilibrium (LD) score regression can be used to estimate the level of overlap between two data sources.

Relationship of one-sample and two-sample Mendelian randomization: populations and samples. In all examples, the green box represents the same underlying population from which samples are drawn; the black circles represent the samples and the text in these summarises the source of association of genetic instrument with exposure (βZX) and association of genetic instrument with outcome (βZY). In one-sample MR (A) where βZX and βZY are estimated within the same population, there may be over-fitting of the data because the predicted (by genetic IV) values of X are then used to predict Y in the same sample. In this study type, weak instrument bias will be expected to bias towards the confounded result. In one-sample MR, it is not necessary to have exposures measured on all sample participants. For expensive exposures, these could be measured in a subsample (B). The properties and sources of bias will be broadly similar to those in (A), where exposures are measured in all participants, but the likelihood of weak instrument bias may be greater. When βZX is obtained in a one-sample MR study but with external weights (i.e., the association magnitudes taken from a GWAS to which the sample being used for the MR did not contribute), as shown in (C), over-fitting of the data is minimised. Ideally, in two-sample MR, both samples are drawn from the same underlying population but there is no overlap of participants between the two samples, as shown in (D). In this situation, data will not be over-fitted and any weak instrument bias would be expected to bias towards the null. As GWAS get larger, and with more cohorts contributing to them, the potential for overlap between samples in summary data two-sample MR becomes increasingly likely, as shown in (E). The more overlap there is between the two samples, the more effects of over-fitting and weak instrument bias become similar to those seen in one-sample MR. In figure (F), the two samples are drawn from two different underlying populations. This might occur when using MR for testing developmental origins, when βZX is estimated in pregnant women and βZY is estimated in their offspring. In that situation, it is important to consider (and ideally test) whether the βZX association in pregnancy is the same as in non-pregnant females and males (i.e., as in the offspring sample). Similarly, when using aggregate data in two-sample MR and when the outcome of interest can only occur in one sex (e.g., cervical or prostate cancer), ideally one would want aggregate βZX estimates to be sex-specific. If that is not possible, then drawing on other external evidence to consider the extent to which βZX is likely to be similar in females and males is important.
Figure 2.6 - Relationship of one-sample and two-sample Mendelian randomization: populations and samples. In all examples, the green box represents the same underlying population from which samples are drawn; the black circles represent the samples and the text in these summarises the source of association of genetic instrument with exposure (βZX) and association of genetic instrument with outcome (βZY). In one-sample MR (A) where βZX and βZY are estimated within the same population, there may be over-fitting of the data because the predicted (by genetic IV) values of X are then used to predict Y in the same sample. In this study type, weak instrument bias will be expected to bias towards the confounded result. In one-sample MR, it is not necessary to have exposures measured on all sample participants. For expensive exposures, these could be measured in a subsample (B). The properties and sources of bias will be broadly similar to those in (A), where exposures are measured in all participants, but the likelihood of weak instrument bias may be greater. When βZX is obtained in a one-sample MR study but with external weights (i.e., the association magnitudes taken from a GWAS to which the sample being used for the MR did not contribute), as shown in (C), over-fitting of the data is minimised. Ideally, in two-sample MR, both samples are drawn from the same underlying population but there is no overlap of participants between the two samples, as shown in (D). In this situation, data will not be over-fitted and any weak instrument bias would be expected to bias towards the null. As GWAS get larger, and with more cohorts contributing to them, the potential for overlap between samples in summary data two-sample MR becomes increasingly likely, as shown in (E). The more overlap there is between the two samples, the more effects of over-fitting and weak instrument bias become similar to those seen in one-sample MR. In figure (F), the two samples are drawn from two different underlying populations. This might occur when using MR for testing developmental origins, when βZX is estimated in pregnant women and βZY is estimated in their offspring. In that situation, it is important to consider (and ideally test) whether the βZX association in pregnancy is the same as in non-pregnant females and males (i.e., as in the offspring sample). Similarly, when using aggregate data in two-sample MR and when the outcome of interest can only occur in one sex (e.g., cervical or prostate cancer), ideally one would want aggregate βZX estimates to be sex-specific. If that is not possible, then drawing on other external evidence to consider the extent to which βZX is likely to be similar in females and males is important.

References

Other terms in 'Sources of bias and limitations in MR':