Title



     mreggerplot -- Scatter plot for MR-Egger type models



Syntax



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



    options               Description
    ----------------------------------------------------------------------------------------------
      egger               MR-Egger estimator for fitted line
      ellipses            marker confidence intervals as ellipses
      errorbars           marker confidence intervals as capped lines
      fe                  fixed effect standard errors for fitted line
      gpci                CIs around G-P associations
      ivw                 IVW line (default is MR-Egger)
      legend(string)      legend options
      level(#)            set confidence level; default is level(95)
      linetop             draw fitted line (and CI) on top of points
      mleglabel(string)   Label for data points in legend; default is Instruments
      median              median estimator for fitted line
      mlabel(string)      variable containing marker labels
      nolci               do not plot confidence interval around fitted line
      noline              do not plot fitted line
      nomcis              do not plot confidence intervals around markers
      penweighted         penalized weighted estimator
      re                  random effect estimator for fitted line
      recons              random intercept for fitted line
      reslope             random slope for fitted line
      reps(#)             number of bootstrap replications to obtain standard error
      seed(#)             seed for random number generator for bootstrapping to obtain standard
                            error
      weighted            weighted median estimator
      wmarkers            weighted markers
      *                   Other options passed to the twoway plot



Description



    mreggerplot plots a scatter plot for MR-Egger type models



    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.



Examples



    Using the data provided by Do et al. (2013) recreate Bowden et al. (2016), Figure 4, LDL-c
    "All genetic variants" (plot in row 2, column 1).



    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)



    Scatter plot of MR-Egger model
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1



    Scatter plot of MR-Egger model labelling outlying genotypes
        . gen mlabvar = rsid if abs(ldlcbeta) > .3
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, mlab(mlabvar) mlabsize(vsmall)
            mlabp(7) mlabc(gs0)



    Scatter plot of MR-Egger model with genotype-phenotype CIs
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, gpci



    Scatter plot of MR-Egger model specifying own titles and legend label
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, title(Investigating pleiotropy:
            MR-Egger model fit) xtitle(Genotype-LDLC associations) ytitle(Genotype-CHD
            associations) mleglabel(Genotypes)



    Scatter plot of IVW model
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, ivw



    Scatter plot of unweighted median model
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, median



    Scatter plot of MR-Egger model using ellipses around points
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, ellipses



    Scatter plot of MR-Egger model adding an IVW line (with legend entry) for comparison
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1
        . mregger chdbeta ldlcbeta [aw=1/(chdse^2)] if sel1==1, ivw
        . addplot : function _b[ldlcbeta]*x if sel1==1, range(0 0.5) lc(gs0) lp(longdash)
            lw(vthin) legend(order(5 "Instruments" 4 "95% CIs" 3 "MR-Egger" 2 "MR-Egger 95% CI" 6
            "IVW") rows(1) size(vsmall))



    Scatter plot of MR-Egger model adding weighted median and modal lines (with legend entries)
    for comparison
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1
        . mrmedian chdbeta chdse ldlcbeta ldlcse if sel1==1, weighted
        . addplot : function _b[beta]*x if sel1==1, range(0 0.5) lc(gs0) lp(shortdash) lw(vthin)
        . mrmodal chdbeta chdse ldlcbeta ldlcse if sel1==1, phi(.25)
        . addplot : function _b[beta]*x if sel1==1, range(0 0.5) lc(gs0) lp(longdash)
            legend(order(5 "Instruments" 4 "95% CIs" 3 "MR-Egger" 2 "MR-Egger 95% CI" 6 "Weighted
            median" 7 "Modal") rows(1) si(vsmall) symx(*.5))



    Scatter plot without any fitted lines
        . mreggerplot chdbeta chdse ldlcbeta ldlcse if sel1==1, ivw noline nolci gpci
            legend(order(4 "Instruments" 3 "95% CI") size(vsmall) rows(1))



References



    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



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



Author



INCLUDE help mrrobust-author