diff --git a/DESCRIPTION b/DESCRIPTION
index 368900a..3173734 100644
--- a/DESCRIPTION
+++ b/DESCRIPTION
@@ -1,7 +1,7 @@
 Package: BoolTraineR
 Type: Package
 Title: Tools For Training and Analysing Asynchronous Boolean Models
-Version: 1.1.1
+Version: 1.1.2
 Date: 2015-10-22
 Author: Chee Yee Lim
 Maintainer: Chee Yee Lim <cyl49@cam.ac.uk>
@@ -22,6 +22,7 @@ LinkingTo: Rcpp
 License: GPL-3
 LazyData: true
 Suggests:
+    bnlearn (>= 3.8.1),
     knitr,
     rmarkdown
 VignetteBuilder: knitr
diff --git a/NAMESPACE b/NAMESPACE
index e2ec62b..6495ba8 100644
--- a/NAMESPACE
+++ b/NAMESPACE
@@ -9,8 +9,6 @@ export(calc_roc)
 export(compress_bmodel)
 export(decompress_bmodel)
 export(df_to_bm)
-export(gen_randata)
-export(gen_randata_bn)
 export(gen_two_rmodel)
 export(gen_two_rmodel_dag)
 export(get_encodings)
diff --git a/R/data_desc.R b/R/data_desc.R
index 2fbbb5c..397251b 100644
--- a/R/data_desc.R
+++ b/R/data_desc.R
@@ -96,15 +96,41 @@ NULL
 #' @usage data(wilson_raw_rnaseq)
 NULL
 
-#' @title Example Boolean Models used in the vignette
+#' @title Example Boolean Model used in the vignette
 #' 
 #' @description
-#' 3 Boolean models used in the examples of the vignette.
+#' A Boolean model used in the examples of the vignette.
 #' 
 #' @format
 #' Each Boolean model is a BoolModel object.
 #' 
 #' @docType data
-#' @name example_models
+#' @name emodel1
 #' @usage data(example_models)
-NULL
\ No newline at end of file
+NULL
+
+#' @title Example Boolean Model used in the vignette
+#' 
+#' @description
+#' A Boolean model used in the examples of the vignette.
+#' 
+#' @format
+#' Each Boolean model is a BoolModel object.
+#' 
+#' @docType data
+#' @name emodel2
+#' @usage data(example_models)
+NULL
+
+#' @title Example Boolean Model used in the vignette
+#' 
+#' @description
+#' A Boolean model used in the examples of the vignette.
+#' 
+#' @format
+#' Each Boolean model is a BoolModel object.
+#' 
+#' @docType data
+#' @name emodel3
+#' @usage data(example_models)
+NULL
diff --git a/R/initialisation.R b/R/initialisation.R
index 56ac7e5..e3704af 100644
--- a/R/initialisation.R
+++ b/R/initialisation.R
@@ -12,8 +12,6 @@
 #' @export
 initialise_raw_data = function(x, data_type='qpcr', uni_thre=0.2, scale=T)
 {
-  require(diptest)
-  
   #(1) Convert negative to positive values.
   if(min(x)<0)
   {
@@ -53,7 +51,7 @@ initialise_raw_data = function(x, data_type='qpcr', uni_thre=0.2, scale=T)
   for(i in 1:ncol(x))
   {
     #Perform unimodality test for each gene.
-    uni_test = dip.test(x[,i])$p.value
+    uni_test = diptest::dip.test(x[,i])$p.value
     
     if(uni_test > uni_thre)
     {
diff --git a/R/methods.R b/R/methods.R
index 5a132a1..ffef26e 100644
--- a/R/methods.R
+++ b/R/methods.R
@@ -66,6 +66,7 @@ writeBM = function(bmodel, file, gene.names=F, rownames=F)
 #' 
 #' @description
 #' This method plots the network underlying Boolean models by using igraph for quick visualisation.
+#' Require igraph.
 #' 
 #' @param bmodel S4 BoolModel object.
 #' @param makePlot logical. Whether to make plot or just return the object. Default to T.
@@ -74,29 +75,27 @@ writeBM = function(bmodel, file, gene.names=F, rownames=F)
 #' @export
 plotBM = function(bmodel, makePlot=T, ...)
 {
-  require(igraph)
-  
   #Convert to amat.
   am = bm_to_amat(bmodel)
   
   #Convert into a graph.
-  g = graph.adjacency(am, mode='directed', weighted=T)
+  g = igraph::graph.adjacency(am, mode='directed', weighted=T)
   
   #Setup edge colour for plotting.
   #Activation = black, inhibition = red
-  E(g)$color = sapply(E(g)$weight, function(x) ifelse(x==1, 'black', 'red'))
+  igraph::E(g)$color = sapply(igraph::E(g)$weight, function(x) ifelse(x==1, 'black', 'red'))
   
   #Setup other colours.
-  V(g)$frame.color = "white"
-  V(g)$color = rgb(255, 165, 0, 200, maxColorValue = 255)
+  igraph::V(g)$frame.color = "white"
+  igraph::V(g)$color = rgb(255, 165, 0, 200, maxColorValue = 255)
   
   #Setup vertex font size.
-  V(g)$label.cex = 1.5
+  igraph::V(g)$label.cex = 1.5
   
   if(makePlot)
   {
     #Make the plot.
-    plot(g, layout=layout_in_circle, ...)
+    igraph::plot.igraph(g, layout=igraph::layout_in_circle, ...)
   }
   
   invisible(g)
@@ -265,7 +264,7 @@ bm_to_df = function(bmodel)
 #' This method converts a data frame into a BoolModel object.
 #' Note that the model should only has 1 NOT operator. More than 1 is STRICTLY NOT allowed.
 #' 
-#' @param df data frame with 2 columns, targets and factors
+#' @param in_df data frame with 2 columns, targets and factors
 #' 
 #' @export
 df_to_bm = function(in_df)
diff --git a/R/rand_model.R b/R/rand_model.R
index db07224..dd2ac8c 100644
--- a/R/rand_model.R
+++ b/R/rand_model.R
@@ -152,8 +152,6 @@ gen_singlerule = function(x, np, tar_ind, and_bool, self_loop=F)
 #' The number of terms in a function for a gene is modelled by power-law distribution.
 gen_one_rmodel = function(var, mvar=length(var), and_bool, self_loop=F)
 {
-  require(poweRlaw)
-  
   if(mvar > length(var))
   {
     mvar=length(var)
@@ -172,140 +170,6 @@ gen_one_rmodel = function(var, mvar=length(var), and_bool, self_loop=F)
   return(bmodel)
 }
 
-#' @title Generate sets of random data
-#' 
-#' @description
-#' This function generates specified sets of random data, which include initial states, two Boolean models (starting and destination), and continuous+discrete data of the destination model.
-#' 
-#' @param n integer. Number of sets of random data required.
-#' @param steps integer. Specify the number of steps between the two Boolean models. If steps=0, give completely random starting model.
-#' @param num_genes integer. Number of genes in the Boolean models.
-#' @param max_varperrule integer. Maximum number of terms per rule (combining both act and inh rule). Note that this number must not be smaller than number of variables. Default to 6.
-#' @param and_bool logical. Indicate whether to consider AND terms.
-#' 
-#' @export
-gen_randata = function(n, steps, num_genes, max_varperrule, and_bool) 
-{
-  var = paste('v', seq(1, num_genes), 's', sep='')
-  
-  output = list()
-  for(i in 1:n)
-  {
-    istate = rbinom(length(var), 1, 0.5)
-    #Getting a random initial state.
-    while(mean(istate) > 0.9 | mean(istate) < 0.1) #do not want initial state that is too homogenous.
-    {
-      istate = rbinom(length(var), 1, 0.5)
-    }
-    istate = data.frame(matrix(istate, nrow=1))
-    colnames(istate) = var
-    istate = initialise_data(istate)
-    
-    #Setting the starting and ending models.
-    bmodel_pair = gen_two_rmodel(var, steps, max_varperrule, and_bool)
-    bmodel_start = bmodel_pair[[1]]
-    bmodel_end = bmodel_pair[[2]]
-    
-    #Getting the simulated data of bmodel_end.
-    ddata = simulate_model(bmodel_end, istate)
-    
-    #Getting parameters from real data.
-    #real_param = param_bimodal(wilson_raw_data, data_type='qpcr') #only need to run once.
-    data(real_param)
-    
-    #Converting binary values to continuous values.
-    cdata = bin_to_real(ddata, real_param)
-    
-    stopifnot(dim(ddata)==dim(cdata))
-    rownames(cdata) = rownames(ddata)
-    colnames(cdata) = colnames(ddata)
-    
-    out_entry = list(istate=istate, bmodel_start=bmodel_start, bmodel_end=bmodel_end, ddata=ddata, cdata=cdata)
-    
-    output = c(output, list(out_entry))
-  }
-  
-  return(output)
-}
-
-#' @title Generate sets of random data
-#' 
-#' @description
-#' (Requires bnlearn) This function generates specified sets of random data, which include initial states, two Boolean models (starting and destination), and continuous+discrete data of the destination model.
-#' 
-#' @param n integer. Number of sets of random data required.
-#' @param steps integer. Specify the number of steps between the two Boolean models. If steps=0, give completely random starting model.
-#' @param num_genes integer. Number of genes in the Boolean models.
-#' @param max_varperrule integer. Maximum number of terms per rule (combining both act and inh rule). Note that this number must not be smaller than number of variables. Default to 6.
-#' 
-#' @export
-gen_randata_bn = function(n, steps, num_genes, max_varperrule) 
-{
-  if (!requireNamespace("bnlearn", quietly = TRUE)) {
-    stop("Package bnlearn needed for this function to work. Please install it.",
-         call. = FALSE)
-  }
-  require('bnlearn')
-  
-  var = paste('v', seq(1, num_genes), 's', sep='')
-  
-  output = list()
-  for(i in 1:n)
-  {
-    istate = rbinom(length(var), 1, 0.5)
-    #Getting a random initial state.
-    while(mean(istate) > 0.9 | mean(istate) < 0.1) #do not want initial state that is too homogenous.
-    {
-      istate = rbinom(length(var), 1, 0.5)
-    }
-    istate = data.frame(matrix(istate, nrow=1))
-    colnames(istate) = var
-    istate = initialise_data(istate)
-    
-    #     #Generate random binary values.
-    #     bn_data = matrix(replicate(num_genes * 100, round(runif(1))), ncol=num_genes)
-    #     
-    #     #Getting parameters from real data.
-    #     #real_param = param_bimodal(wilson_raw_data, data_type='qpcr') #only need to run once.
-    #     data(real_param)
-    #     
-    #     #Converting binary values to continuous values.
-    #     bn_data = data.frame(bin_to_real(bn_data, real_param))
-    #     colnames(bn_data) = var
-    
-    #Setting the starting and ending models.
-    bn_model_pair = gen_two_rmodel_dag(var, steps, max_varperrule)
-    bn_model_start = bn_model_pair[[1]]
-    bn_model_end = bn_model_pair[[2]]
-    
-    #Converting the starting and ending models from bn objects to BoolModel objects.
-    bmodel_start = amat_to_bm(amat(bn_model_start))
-    bmodel_end = amat_to_bm(amat(bn_model_end))
-    
-    #Getting the simulated data of bmodel_end.
-    ddata = simulate_model(bmodel_end, istate)
-    
-    #Getting parameters from real data.
-    #real_param = param_bimodal(wilson_raw_data, data_type='qpcr') #only need to run once.
-    data(real_param)
-    
-    #Converting binary values to continuous values.
-    cdata = bin_to_real(ddata, real_param)
-    
-    stopifnot(dim(ddata)==dim(cdata))
-    rownames(cdata) = rownames(ddata)
-    colnames(cdata) = colnames(ddata)
-    
-    out_entry = list(istate=istate, bmodel_start=bmodel_start, bmodel_end=bmodel_end, 
-                     ddata=ddata, cdata=cdata,
-                     bn_model_start=bn_model_start, bn_model_end=bn_model_end)
-    
-    output = c(output, list(out_entry))
-  }
-  
-  return(output)
-}
-
 #' @title Generate two random DAG Boolean models with a specified number of steps apart
 #' 
 #' @description
@@ -321,107 +185,110 @@ gen_randata_bn = function(n, steps, num_genes, max_varperrule)
 #' @export
 gen_two_rmodel_dag = function(var, steps, mvar=length(var), in_amat=NULL, acyclic=T)
 { 
-  require(bnlearn)
-  
-  if(steps==0)
-  {
-    start_model = random.graph(var, method = "melancon", max.degree=mvar)
-    end_model = random.graph(var, method = "melancon", max.degree=mvar)
+  if(!requireNamespace('bnlearn', quietly = TRUE)) {
+    stop('Requires bnlearn package to run this function.')
   } else
   {
-    if(is.null(in_amat))
+    if(steps==0)
     {
-      #(3) Get first model.
-      start_model = random.graph(var, method = "melancon", max.degree=mvar)
+      start_model = bnlearn::random.graph(var, method = "melancon", max.degree=mvar)
+      end_model = bnlearn::random.graph(var, method = "melancon", max.degree=mvar)
     } else
     {
-      start_model = empty.graph(var)
-      if(acyclic)
+      if(is.null(in_amat))
       {
-        amat(start_model) = in_amat
+        #(3) Get first model.
+        start_model = bnlearn::random.graph(var, method = "melancon", max.degree=mvar)
       } else
       {
-        amat(start_model, ignore.cycles=T) = in_amat
+        start_model = bnlearn::empty.graph(var)
+        if(acyclic)
+        {
+          bnlearn::amat(start_model) = in_amat
+        } else
+        {
+          bnlearn::amat(start_model, ignore.cycles=T) = in_amat
+        }
       }
-    }
-    
-    cur_ite = 1
-    max_ite = 100
-    out_break = F
-    while(cur_ite < max_ite & !out_break)
-    {
-      #(4) Get second model of a specified number of steps away.
-      end_model = start_model
-      memory_ind = c()
-      #memory_ind = apply(which(t(amat(end_model))==1, arr.ind = T), 1, function(x) paste(x, collapse=',')) #no undirected arcs.
-      #memory_ind = c(memory_ind, paste(1:nrow(tmp), 1:ncol(tmp), sep=',')) #no self_loops.
-      #start_time = proc.time()
-      while(TRUE)
+      
+      cur_ite = 1
+      max_ite = 100
+      out_break = F
+      while(cur_ite < max_ite & !out_break)
       {
-        tryCatch({
-          #used_time = proc.time() - start_time
-          #used_time = as.numeric(used_time)[3]
-          
-          tmp_graph = empty.graph(var)
-          tmp = amat(end_model)
-          
-          x_ind = sample(1:nrow(tmp),1)
-          y_ind = sample(1:ncol(tmp),1)
-          
-          while(paste(x_ind, y_ind, collapse=',') %in% memory_ind)
-          {
+        #(4) Get second model of a specified number of steps away.
+        end_model = start_model
+        memory_ind = c()
+        #memory_ind = apply(which(t(amat(end_model))==1, arr.ind = T), 1, function(x) paste(x, collapse=',')) #no undirected arcs.
+        #memory_ind = c(memory_ind, paste(1:nrow(tmp), 1:ncol(tmp), sep=',')) #no self_loops.
+        #start_time = proc.time()
+        while(TRUE)
+        {
+          tryCatch({
+            #used_time = proc.time() - start_time
+            #used_time = as.numeric(used_time)[3]
+            
+            tmp_graph = bnlearn::empty.graph(var)
+            tmp = bnlearn::amat(end_model)
+            
             x_ind = sample(1:nrow(tmp),1)
             y_ind = sample(1:ncol(tmp),1)
-          }
-          
-          memory_ind = c(memory_ind, paste(x_ind, y_ind, collapse=','))
-          if(length(memory_ind) == nrow(tmp)*ncol(tmp)) #break loop, get a new starting model and try again.
-          {
-            #cat(sprintf('For step %s, no other possible model exists.', steps))
-            break
-          }
-          
-          #Change one step at a time.
-          if(tmp[x_ind,y_ind]==0)
-          {
-            tmp[x_ind,y_ind] = 1
-          } else
-          {
-            tmp[x_ind,y_ind] = 0
-          }
-          
-          if(acyclic)
-          {
-            amat(tmp_graph) = tmp
-          } else
-          {
-            amat(tmp_graph, ignore.cycles=T) = tmp
-          }
-          
-          end_model = tmp_graph
-          
-          if(sum(abs(amat(start_model)-amat(tmp_graph))) == steps)
-          {
-            if(nrow(undirected.arcs(tmp_graph))==0) #cannot have undirected arcs in the models.
+            
+            while(paste(x_ind, y_ind, collapse=',') %in% memory_ind)
+            {
+              x_ind = sample(1:nrow(tmp),1)
+              y_ind = sample(1:ncol(tmp),1)
+            }
+            
+            memory_ind = c(memory_ind, paste(x_ind, y_ind, collapse=','))
+            if(length(memory_ind) == nrow(tmp)*ncol(tmp)) #break loop, get a new starting model and try again.
             {
-              out_break = T
+              #cat(sprintf('For step %s, no other possible model exists.', steps))
               break
             }
-          }
-        }, error=function(e){},
-        finally={})
+            
+            #Change one step at a time.
+            if(tmp[x_ind,y_ind]==0)
+            {
+              tmp[x_ind,y_ind] = 1
+            } else
+            {
+              tmp[x_ind,y_ind] = 0
+            }
+            
+            if(acyclic)
+            {
+              bnlearn::amat(tmp_graph) = tmp
+            } else
+            {
+              bnlearn::amat(tmp_graph, ignore.cycles=T) = tmp
+            }
+            
+            end_model = tmp_graph
+            
+            if(sum(abs(bnlearn::amat(start_model)-bnlearn::amat(tmp_graph))) == steps)
+            {
+              if(nrow(bnlearn::undirected.arcs(tmp_graph))==0) #cannot have undirected arcs in the models.
+              {
+                out_break = T
+                break
+              }
+            }
+          }, error=function(e){},
+          finally={})
+        }
+        
+        cur_ite = cur_ite + 1
       }
       
-      cur_ite = cur_ite + 1
+      if(cur_ite == max_ite)
+      {
+        stop(sprintf('For step %s, no other possible model exists.', steps))
+      }
     }
     
-    if(cur_ite == max_ite)
-    {
-      stop(sprintf('For step %s, no other possible model exists.', steps))
-    }
+    return(list(start_model, end_model))
   }
-  
-  return(list(start_model, end_model))
 }
 
 #' @title Generate two random Boolean models with a specified number of steps apart
diff --git a/R/score_calculation.R b/R/score_calculation.R
index 5501305..41689c2 100644
--- a/R/score_calculation.R
+++ b/R/score_calculation.R
@@ -62,7 +62,7 @@ m_score = function(x, bmodel, max_varperrule, detail=F)
 #' 
 #' @param bmodel S4 BoolModel object. Model to be evaluated.
 #' @param istate data frame. Must have only 1 row, which represents 1 initial state.
-#' @param data matrix. Represents the expression data df.
+#' @param fcdata matrix. Represents the expression data df.
 #' @param overlap_gene character vector. Specify which genes are present in both model and data inputs.
 #' @param max_varperrule integer. Maximum number of terms per rule (combining both act and inh rule). Note that this number must not be smaller than number of variables. Default to 6.
 #' @param detail logical. Whether to give more details in score calculation. Default to FALSE.
diff --git a/R/search.R b/R/search.R
index 4812212..cde2ccf 100644
--- a/R/search.R
+++ b/R/search.R
@@ -250,7 +250,10 @@ model_train = function(edata, bmodel=NULL, istate=NULL, max_varperrule=6, and_bo
     }
     
     vcat('Stage 1: Exploring neighbouring models.\n', verbose)
-    mod_model = sample(unlist(minmod_model(mod_model[[i]], overlap_gene=overlap_gene)), 1)
+    mod_model = foreach(i=1:length(mod_model)) %dopar% {
+      c(mod_model[[i]], unlist(minmod_model(mod_model[[i]], overlap_gene=overlap_gene)))
+    }
+    mod_model = unlist(mod_model)
     
     vcat(sprintf('Total neighbouring models: %s.\n', length(mod_model)), verbose)
     if(length(mod_model)>1000000)
diff --git a/booltrainer.Rproj b/booltrainer.Rproj
deleted file mode 100644
index 8ecf4c0..0000000
--- a/booltrainer.Rproj
+++ /dev/null
@@ -1,18 +0,0 @@
-Version: 1.0
-
-RestoreWorkspace: Default
-SaveWorkspace: Default
-AlwaysSaveHistory: Default
-
-EnableCodeIndexing: Yes
-UseSpacesForTab: Yes
-NumSpacesForTab: 2
-Encoding: UTF-8
-
-RnwWeave: knitr
-LaTeX: pdfLaTeX
-
-BuildType: Package
-PackageUseDevtools: Yes
-PackageInstallArgs: --no-multiarch --with-keep.source
-PackageRoxygenize: rd,namespace
diff --git a/man/calc_mscore.Rd b/man/calc_mscore.Rd
index e9ddb87..a1224c5 100644
--- a/man/calc_mscore.Rd
+++ b/man/calc_mscore.Rd
@@ -11,13 +11,13 @@ calc_mscore(bmodel, istate, fcdata, overlap_gene, max_varperrule, detail = F)
 
 \item{istate}{data frame. Must have only 1 row, which represents 1 initial state.}
 
+\item{fcdata}{matrix. Represents the expression data df.}
+
 \item{overlap_gene}{character vector. Specify which genes are present in both model and data inputs.}
 
 \item{max_varperrule}{integer. Maximum number of terms per rule (combining both act and inh rule). Note that this number must not be smaller than number of variables. Default to 6.}
 
 \item{detail}{logical. Whether to give more details in score calculation. Default to FALSE.}
-
-\item{data}{matrix. Represents the expression data df.}
 }
 \description{
 This function calculates a score for a Boolean model wrt to a dataset.
diff --git a/man/df_to_bm.Rd b/man/df_to_bm.Rd
index 14944bc..04923a2 100644
--- a/man/df_to_bm.Rd
+++ b/man/df_to_bm.Rd
@@ -7,7 +7,7 @@
 df_to_bm(in_df)
 }
 \arguments{
-\item{df}{data frame with 2 columns, targets and factors}
+\item{in_df}{data frame with 2 columns, targets and factors}
 }
 \description{
 This method converts a data frame into a BoolModel object.
diff --git a/man/example_models.Rd b/man/emodel1.Rd
similarity index 58%
rename from man/example_models.Rd
rename to man/emodel1.Rd
index 7355b88..b51d416 100644
--- a/man/example_models.Rd
+++ b/man/emodel1.Rd
@@ -1,14 +1,14 @@
 % Generated by roxygen2 (4.1.1): do not edit by hand
 % Please edit documentation in R/data_desc.R
 \docType{data}
-\name{example_models}
-\alias{example_models}
-\title{Example Boolean Models used in the vignette}
+\name{emodel1}
+\alias{emodel1}
+\title{Example Boolean Model used in the vignette}
 \format{Each Boolean model is a BoolModel object.}
 \usage{
 data(example_models)
 }
 \description{
-3 Boolean models used in the examples of the vignette.
+A Boolean model used in the examples of the vignette.
 }
 
diff --git a/man/emodel2.Rd b/man/emodel2.Rd
new file mode 100644
index 0000000..7c26d86
--- /dev/null
+++ b/man/emodel2.Rd
@@ -0,0 +1,14 @@
+% Generated by roxygen2 (4.1.1): do not edit by hand
+% Please edit documentation in R/data_desc.R
+\docType{data}
+\name{emodel2}
+\alias{emodel2}
+\title{Example Boolean Model used in the vignette}
+\format{Each Boolean model is a BoolModel object.}
+\usage{
+data(example_models)
+}
+\description{
+A Boolean model used in the examples of the vignette.
+}
+
diff --git a/man/emodel3.Rd b/man/emodel3.Rd
new file mode 100644
index 0000000..a68d0ff
--- /dev/null
+++ b/man/emodel3.Rd
@@ -0,0 +1,14 @@
+% Generated by roxygen2 (4.1.1): do not edit by hand
+% Please edit documentation in R/data_desc.R
+\docType{data}
+\name{emodel3}
+\alias{emodel3}
+\title{Example Boolean Model used in the vignette}
+\format{Each Boolean model is a BoolModel object.}
+\usage{
+data(example_models)
+}
+\description{
+A Boolean model used in the examples of the vignette.
+}
+
diff --git a/man/gen_randata.Rd b/man/gen_randata.Rd
deleted file mode 100644
index cdc0960..0000000
--- a/man/gen_randata.Rd
+++ /dev/null
@@ -1,23 +0,0 @@
-% Generated by roxygen2 (4.1.1): do not edit by hand
-% Please edit documentation in R/rand_model.R
-\name{gen_randata}
-\alias{gen_randata}
-\title{Generate sets of random data}
-\usage{
-gen_randata(n, steps, num_genes, max_varperrule, and_bool)
-}
-\arguments{
-\item{n}{integer. Number of sets of random data required.}
-
-\item{steps}{integer. Specify the number of steps between the two Boolean models. If steps=0, give completely random starting model.}
-
-\item{num_genes}{integer. Number of genes in the Boolean models.}
-
-\item{max_varperrule}{integer. Maximum number of terms per rule (combining both act and inh rule). Note that this number must not be smaller than number of variables. Default to 6.}
-
-\item{and_bool}{logical. Indicate whether to consider AND terms.}
-}
-\description{
-This function generates specified sets of random data, which include initial states, two Boolean models (starting and destination), and continuous+discrete data of the destination model.
-}
-
diff --git a/man/gen_randata_bn.Rd b/man/gen_randata_bn.Rd
deleted file mode 100644
index d673660..0000000
--- a/man/gen_randata_bn.Rd
+++ /dev/null
@@ -1,21 +0,0 @@
-% Generated by roxygen2 (4.1.1): do not edit by hand
-% Please edit documentation in R/rand_model.R
-\name{gen_randata_bn}
-\alias{gen_randata_bn}
-\title{Generate sets of random data}
-\usage{
-gen_randata_bn(n, steps, num_genes, max_varperrule)
-}
-\arguments{
-\item{n}{integer. Number of sets of random data required.}
-
-\item{steps}{integer. Specify the number of steps between the two Boolean models. If steps=0, give completely random starting model.}
-
-\item{num_genes}{integer. Number of genes in the Boolean models.}
-
-\item{max_varperrule}{integer. Maximum number of terms per rule (combining both act and inh rule). Note that this number must not be smaller than number of variables. Default to 6.}
-}
-\description{
-(Requires bnlearn) This function generates specified sets of random data, which include initial states, two Boolean models (starting and destination), and continuous+discrete data of the destination model.
-}
-
diff --git a/man/plotBM.Rd b/man/plotBM.Rd
index 5028161..8e7d538 100644
--- a/man/plotBM.Rd
+++ b/man/plotBM.Rd
@@ -15,5 +15,6 @@ plotBM(bmodel, makePlot = T, ...)
 }
 \description{
 This method plots the network underlying Boolean models by using igraph for quick visualisation.
+Require igraph.
 }
 
diff --git a/vignettes/booltrainer.Rmd b/vignettes/booltrainer.Rmd
index 8dfaa71..30bbe4f 100644
--- a/vignettes/booltrainer.Rmd
+++ b/vignettes/booltrainer.Rmd
@@ -6,17 +6,20 @@ output:
     rmarkdown::html_vignette:
         toc: true
         number_sections: true
+vignette: >
+  %\VignetteIndexEntry{Using BoolTraineR to reconstruct asynchronous Boolean models}
+  %\VignetteEngine{knitr::rmarkdown}
+  %\VignetteEncoding{UTF-8}
+---
+
+<!---
     rmarkdown::pdf_document:
         toc: true
         number_sections: true
     rmarkdown::md_document:
         variant: markdown_github
         toc: true
-vignette: >
-  %\VignetteIndexEntry{Using BoolTraineR to reconstruct asynchronous Boolean models}
-  %\VignetteEngine{knitr::rmarkdown}
-  %\VignetteEncoding{UTF-8}
----
+-->
 
 ```{r, echo=FALSE}
 library('BoolTraineR')