Title



     mrfunnel -- Funnel plot for two-sample MR analysis



Syntax



        mrfunnel varname_gd varname_gdse varname_gp varname_gpse [if] [in] [, options]



    options               Description
    ----------------------------------------------------------------------------------------------
      extraplots(string)  extra plots to add to the overall plot
      metric(metric)      scale of y-axis
      noivw               do not plot IVW line
      nomregger           do not plot MR-Egger line
      mrivestsopts(opts)  options passed to mrivests
      scatteropts(opts)   options passed to the scatter command
      xlrange(# #)        the range for the IVW and MR-Egger lines, see twoway_function range()
                            option
      *                   other options passed to twoway



Description



    mrfunnel provides a funnel plot for a two-sample Mendelian randomization analysis.



    There are 3 choices of measures of instrument strength to plot on the y-axis, which are
    specified using the metric option and are described below.
 
    On the plot the MR-Egger estimate is the line with the longer dashes, the IVW estimate is
    shown with the shorter dashes.



    varname_gd is a variable containing the genotype-disease (SNP-outcome) association estimates.



    varname_gdse is a variable containing the genotype-disease (SNP-outcome) association estimate
    standard errors.



    varname_gp is a variable containing the genotype-phenotype (SNP-exposure) association
    estimates.



    varname_gpse is a variable containing the genotype-phenotype (SNP-exposure) association
    estimate standard errors.



Options



    metric(gpbeta|gpbetastd|invse) specifies the metric for the y-axis. Can be one of:
    - gpbeta: the absolute value of the genotype-phenotype (SNP-exposure) estimates,
    - gpbetastd: gpbeta standardised by the genotype-disease (SNP-outcome) standard errors (the
    default),
    - invse: the inverse of the standard errors on the genotype specific IV ratio estimates.



    Note that in Stata 18.0 the default legend position of a twoway plot was changed from the 6
    o'clock position to the 3 o'clock position.  In Stata 18.0 and above mrfunnel resets the
    default back to 6 o'clock (which can be overridden with, for example, legend(pos(12))).



Examples



    Using the data provided by Do et al. (2013) recreate Bowden et al. (2016) Web Figure A2
    (top-right plot, LDL-C with 73 genotypes).



    Setup
        . use https://raw.github.com/remlapmot/mrrobust/master/dodata, clear



    Select observations (p-value with exposure < 10^-8)
        . gen byte sel1 = (ldlcp2 < 1e-8)



    Funnel plot
        . mrfunnel chdbeta chdse ldlcbeta ldlcse if sel1==1



    Without adding the IVW and MR-Egger estimates
        . mrfunnel chdbeta chdse ldlcbeta ldlcse if sel1==1, noivw nomregger



    Using an unstandardised y-axis
        . mrfunnel chdbeta chdse ldlcbeta ldlcse if sel1==1, metric(gpbeta)



    Using inverse IV SEs on the y-axis
        . mrfunnel chdbeta chdse ldlcbeta ldlcse if sel1==1, metric(invse)



    Remove the legend
        . mrfunnel chdbeta chdse ldlcbeta ldlcse if sel1==1, legend(off)



    Extend the IVW and MR-Egger lines to the y-axis limits (as per the original version of this
    command)
        . mrfunnel chdbeta chdse ldlcbeta ldlcse if sel1==1, xlrange(0 10)



References



    Do et al., 2013. Common variants associated with plasma triglycerides and risk for coronary
        artery disease. Nature Genetics. 45, 1345-1352.  DOI



    Bowden J, Davey Smith G, Haycock PC, Burgess S. Consistent estimation in Mendelian
        randomization with some invalid instruments using a weighted median estimator. Genetic
        Epidemiology, 2016, 40, 4, 304-314.  DOI



Author



INCLUDE help mrrobust-author