- Function
**comp_clustering**:- Performs quantitative evaluation of the transitivity assumption using inter-trial dissimilarities for various trial-level aggregate participant and methodological characteristics that may act as effect modifiers.

- Function
**dendro_heatmap**:- Returns the dendrogram with integrated heatmap of the clustered comparisons and trials based on hierarchical agglomerative clustering (performed using the function
**comp_clustering**). The R packages*heatmaply*and*dendextend*have been used.

- Returns the dendrogram with integrated heatmap of the clustered comparisons and trials based on hierarchical agglomerative clustering (performed using the function
- Function
**distr_characteristics**:- It returns violin plots with integrated box plots and dots for quantitative characteristics, and stacked barplots for qualitative characteristics across the observed treatment comparisons. The function can also be used to illustrate the distribution of the characteristics across the clusters defined from
**comp_clustering**.

- It returns violin plots with integrated box plots and dots for quantitative characteristics, and stacked barplots for qualitative characteristics across the observed treatment comparisons. The function can also be used to illustrate the distribution of the characteristics across the clusters defined from
- Function
**miss_characteristics**:- It returns various plots to visualise the missing cases in the characteristics across trials and treatment comparisons.

- Function
**gower_distance**:- It returns the N-by-N matrix on Gower’s dissimilarity coefficient for all pairs of N trials in a network.

- Function
**mcmc_diagnostics**:- returns a bar plot on the ratio of MCMC error to the posterior standard deviation and a bar plot on the Gelman-Rubin R diagnostic. Green bars indicate ratio less than 0.05 and R less than 1.10; otherwise, the bars are red.

- Functions
**baseline_model**,**run_metareg**,**run_model**,**run_nodesplit**,**run_sensitivity**,**run_series_meta**, and**run_ume**:- The corresponding models are updated until convergence via the
*autojags*function of the R package*R2jags*. - The argument
*inits*has been added to allow the user define the initial values for the parameters, following the documentation of the*jags*function in the R package*R2jags*.

- The corresponding models are updated until convergence via the
- Function
**describe_network**:- It reports only the tables with the evidence base information on one outcome. The network plot is not reported (see and use
**netplot**, instead).

- It reports only the tables with the evidence base information on one outcome. The network plot is not reported (see and use
- Function
**netplot**:- Self-created function using the R package
*igraph*. This function creates the network plot.

- Self-created function using the R package

- Function
**baseline_model**:- processes the elements in the argument
*base_risk*for a fixed, random or predicted baseline model and passes the output to run_model or run_metareg to obtain the absolute risks for all interventions. - when a predicted baseline model is conducted, it returns a forest plot with the trial-specific and summary probability of an event for the selected reference intervention.

- processes the elements in the argument
- Function
**forestplot_metareg**:- upgraded plot that presents two forest plots side-by-side: (i) one on estimation and prediction from network meta-analysis and network meta-regression for a selected comparator intervention (allows comparison of these two analyses), and (ii) one on SUCRA values from both analyses. Both forest plots present results from network meta-regression for a selected value of the investigated covariate.

- Function
**league_table_absolute_user**:- (only for binary outcome) yields the same graph with league_table_absolute, but the input is not
*rnmamod*object: the user defines the input and it includes the summary effect and corresponding (credible or confidence) interval for comparisons with a reference intervention. These results stem from a network meta-analysis conducted using another R-package or statistical software.

- (only for binary outcome) yields the same graph with league_table_absolute, but the input is not
- Function
**robustness_index_user**:- calculates the robustness index for a sensitivity analysis performed using the R-package
*netmeta*or*metafor*. The user defines the input and the function returns the robustness index. This function returns the same output with the**robustness_index**function.

- calculates the robustness index for a sensitivity analysis performed using the R-package
- Function
**trans_quality**:- classifies a systematic review with multiple interventions as having low, unclear or high quality regarding the transitivity evaluation based on five quality criteria.

- Typos and links (for functions and packages) in the documentation are corrected.
- Function
**run_model**:- allows the user to define the reference intervention of the network via the argument
*ref*; - (only for binary outcome) estimates the absolute risks for all non-reference interventions using a selected baseline risk for the reference intervention (argument
*base_risk*); - (only for binary outcome) estimates the relative risks and risk difference as functions of the estimated absolute risks.

- allows the user to define the reference intervention of the network via the argument
- Function
**league_table_absolute**:- (only for binary outcome) it presents the absolute risks per 1000 participants in main diagonal, the odds ratio on the upper off-diagonals, and the risk difference per 1000 participants in the lower off-diagonals;
- allow the user to select the interventions to present via the argument
*show*(ideal for very large networks that make the league table cluttered).

- Functions
**league_heatmap**and**league_heatmap_pred**:- allow the user to select the interventions to present via the argument
*show*(ideal for very large networks that make the league table cluttered); - allow the user to illustrate the results of two outcomes for the same model (i.e. via run_model or run_metareg) or the results of two models on the same outcome (applicable for: (i) run_model versus run_metareg, and (ii) run_model versus run_series_meta).

- allow the user to select the interventions to present via the argument
- Functions
**series_meta_plot**and**nodesplit_plot**:- present the extent of heterogeneity in the forest plot of tau using different colours for low, reasonable, fairly high, and fairly extreme tau (the classification has been suggested by Spiegelhalter et al., 2004; ISBN 0471499757).

- First CRAN submission.