cumrank() now normalised tobe between 0 and 1.pd is now specified aspD and should belogical as eitherTRUE (estimates the effective number ofparameters via the Kullback-Leibler divergence) orFALSE(using the pV approximation)R2jags that wasignoring the warmup stage and effectively saving all the simulationsperformed for Monte-Carlo estimation.netmeta added to list of suggested packagesregress.vars argument inmbnma.run().Various sharing assumptions for effects can be specified inregress.effect.dfpoly() can only takenumeric values from set defined in Jansen 2015.calc.edx() to allow easy estimation of differentED values (e.g. ED90 = the dose at which 90% of the maximum response(Emax) is reached)get.relative() now allows simultaneous comparison oftwo models in a single league table - can be used to compare MBNMAmodels with different dose-response functions, or MBNMA and NMA models,or NMA models that assume consistency versus those that use UnrelatedMean Effects.ed50,hill,onset) are now on the natural scale and are assignedtruncated normal default priorsgetjagsdata()fitplot() anddevplot()get.relative() to allow estimation of relativeeffects between any doses of different agents."relative.array" objects generatedbyget.relative().n rather thanN so that datasets can beconsistent with those used inMBNMAtimepredict.mbnma() andget.relative()devdev() for comparing deviance contributionsbetween modelsmbnma.run() are nowgiven asclass("dosefun") and dose-response parameters arespecified within these functions. NOTE: Previous syntax of specifying afunction name as a character (e.g. fun="linear") along withbeta parameters (e.g. mbnma.run(beta.1="rel")) will beremoved in subsequent versions, along with wrapper functions.dloglin())dspline()) (piecewise linearsplines, B-splines, restricted cubic splines, natural splines)dfpoly())link="smd" to allow for analysis usingStandardised Mean Differencescalcom() to guess outcome measure scale for morecareful specification of default priors for SD"mbnma.network" objectmbnma.nodesplit() fixedparams inplot.mbnma.rank() is not a subset ofxoverlay.split() uses full distribution ofE0 rather than summary statisticsmbnma.predict object now contains valuesassigned/estimated forE0 to be used inoverlay.split()plot.nodesplit(),plot.type="forest"plots a single forest plot with results for each node-split comparison,rather than presenting results in panels.summary.mbnma.network() returns valid minimumdoses per agentparallel=TRUE andadded a warning whenpd is set to"pd.kl" or"popt" for these models.summary() for multiple dose-response function modelsfun="rcs") inmbnma.run()mbnma.run() to allow relaxing of the consistencyassumption. This can be used to test its validity.cumrank() added for cumulative ranking plots. Alsocalculates SUCRA values for each agent and dose-response parameterautojags options added formbnma.run() toallow users to run models until they converge (convergence defined byRhat)rank.mbnma() also calculates cumulative rankingprobabilities and stores them incum.matrixgetjagsdata() containsstudyIDand has been added tombnma objectsdevplot() andfitplot()plot.nodesplit() scales y-axis if density is >50times larger in panel with highest density than in panel with lowestdensity. This improves legibility of the graph.class("nodesplit")mbnma.nodesplit() includes potential splits viadose-response curve and direct and indirect evidence contributions arecalculated simultaneously in the same model.mbnma.nodesplit() andnma.nodesplit()plot.mbnma.network()psoriasis andssri datasets topackagecrayon package to neaten printed consoleoutputsfun inmbnma.run()) so that multiple functions can be modelledsimultaneously. Some downstream package functions still may not yet workwith these models though.mbnma.network objects returned fromplot.mbnma.network now have specific igraph attributesassigned to them, which can be easily changed by the user.user.fun now takes a formula as an argument (forexample~ (beta.1 * dose) + (beta.2 * dose^2)) rather thana string.plot.mbnma.network() now uses alayoutargument that takes an igraph layout function instead oflayout_in_circle (which was a logical argument). Thisallows any igraph layout to be plotted rather than just a circle(e.g. igraph::as_star())if {class(x)=="matrix"} statements toif {is.matrix(x)} to address R development changespd="plugin"), or Kullback-Leibler divergence(pd="pd.kl")parallel=TRUE inmbnma.run() (or wrapper functions) now properly runs JAGSin parallel on multiple cores.mbnma.network in their output rather than just treatmentand agent names.nma.nodesplit() that prevented themodel running if disconnected treatments were included in the analysis(drop.discon=FALSE)Welcome to MBNMAdose. Ready for release into the world. I hope it canbe of service to you! For time-course MBNMA, also check out the sisterpackage, MBNMAtime.