diff --git a/app.py b/app.py
index ef1b8bc4..86008af0 100644
--- a/app.py
+++ b/app.py
@@ -46,7 +46,8 @@
 
 # ----------- Figures ---------------------
 fig1 = make_map(df_tidy, df_tidy_fatalities)
-fig2 = make_timeplot(df, df_prediction)
+fig2 = make_timeplot(df, df_prediction, countries=['France', 'Italy', 'Spain'])
+fig_store = make_timeplot(df, df_prediction)
 
 # ------------ Markdown text ---------------
 # maybe later we can break the text in several parts
@@ -97,7 +98,7 @@
             ],
             className="pure-u-1 pure-u-lg-1-2 pure-u-xl-8-24",
             ),
-        dcc.Store(id='store', data=[fig2, initial_indices]),
+        dcc.Store(id='store', data=[fig_store, initial_indices]),
         html.Div([
             dash_table.DataTable(
                 id='table',
@@ -165,21 +166,12 @@
 # in order to transform the app into static html pages
 # javascript functions are defined in assets/callbacks.js
 
-app.clientside_callback(
-    ClientsideFunction(
-        namespace='clientside2',
-        function_name='get_store_data'
-    ),
-    output=Output('plot', 'figure'),
-    inputs=[Input('store', 'data')])
-
-
 app.clientside_callback(
     ClientsideFunction(
         namespace='clientside',
         function_name='update_store_data'
     ),
-    output=Output('store', 'data'),
+    output=Output('plot', 'figure'),
     inputs=[
         Input('table', "data"),
         Input('table', "selected_rows")],
@@ -203,6 +195,5 @@
     )
 
 
-
 if __name__ == '__main__':
     app.run_server(debug=debug)
diff --git a/assets/callbacks.js b/assets/callbacks.js
index 98576bf9..602c14d8 100644
--- a/assets/callbacks.js
+++ b/assets/callbacks.js
@@ -3,11 +3,6 @@ if (!window.dash_clientside) {
 }
 
 
-window.dash_clientside.clientside2 = {
-    get_store_data: function(data) {
-        return data[0];
-    }
-};
 
 window.dash_clientside.clientside3 = {
     update_table: function(clickdata, selecteddata, table_data, selectedrows, store) {
@@ -51,27 +46,26 @@ window.dash_clientside.clientside3 = {
 
     
 window.dash_clientside.clientside = {
-    update_store_data: function(selectedrows, rows, store) {
+    update_store_data: function(rows, selectedrows, store) {
 	var fig = store[0];
 	if (!rows) {
            throw "Figure data not loaded, aborting update."
        }
-	var new_fig = {...fig};
+	var new_fig = {};
+	new_fig['data'] = [];
+	new_fig['layout'] = fig['layout'];
 	
 	var countries = [];
-	for (i = 0; i < rows.length; i++) {
-	    countries.push(selectedrows[rows[i]]["country_region"]);
+	for (i = 0; i < selectedrows.length; i++) {
+	    countries.push(rows[selectedrows[i]]["country_region"]);
 	}
-	for (i = 0; i < new_fig['data'].length; i++) {
-	    var name = new_fig['data'][i]['name'];
+	for (i = 0; i < fig['data'].length; i++) {
+	    var name = fig['data'][i]['name'];
 	    if (countries.includes(name) || countries.includes(name.substring(1))){
-		new_fig['data'][i]['visible'] = true;
-	    }
-	    else{
-		new_fig['data'][i]['visible'] = false;
+		new_fig['data'].push(fig['data'][i]);
 	    }
 	}
-        return [new_fig, store[1]];
+        return new_fig;
     }
 };
 
diff --git a/make_figures.py b/make_figures.py
index 8bb4dfb3..201b18cf 100644
--- a/make_figures.py
+++ b/make_figures.py
@@ -57,7 +57,7 @@ def make_map(df, df_fatalities):
     return fig
 
 
-def make_timeplot(df_measure, df_prediction):
+def make_timeplot(df_measure, df_prediction, countries=None):
     """
     Build figure showing evolution of number of cases vs. time for all countries.
     The visibility of traces is set to 0 so that the interactive app will
@@ -82,6 +82,8 @@ def make_timeplot(df_measure, df_prediction):
     hovertemplate_measure = '<b>%{meta}</b><br>%{x}<br>%{y:.0f} per Million<extra></extra>'
     hovertemplate_prediction = '<b>%{meta}<br>prediction</b><br>%{x}<br>%{y:.0f} per Million<extra></extra>'
     for i, country in enumerate(df_measure_confirmed.columns):
+        if countries and country[1] not in countries:
+            continue
         fig.add_trace(go.Scatter(x=df_measure_confirmed.index,
                                  y=df_measure_confirmed[country],
                                  name=country[1], mode='markers+lines',
@@ -90,7 +92,7 @@ def make_timeplot(df_measure, df_prediction):
                                  line_color=colors[i%n_colors],
                                  meta=country[1],
                                  hovertemplate=hovertemplate_measure,
-                                 visible=False))
+                                 visible=True))
     prediction = df_prediction['prediction']
     upper_bound = df_prediction['upper_bound']
     lower_bound = df_prediction['lower_bound']
@@ -101,6 +103,8 @@ def make_timeplot(df_measure, df_prediction):
     lower_bound = normalize_by_population_wide(lower_bound)
     lower_bound *= 1e6
     for i, country in enumerate(prediction.columns):
+        if countries and country[1] not in countries:
+            continue
         # Do not plot predictions for a country with less than 50 cases
         if df_measure_confirmed[country][-1] < 50:
             continue
@@ -112,14 +116,14 @@ def make_timeplot(df_measure, df_prediction):
                                  showlegend=False,
                                  meta=country[1],
                                  hovertemplate=hovertemplate_prediction,
-                                 visible=False))
+                                 visible=True))
         fig.add_trace(go.Scatter(x=upper_bound.index,
                                  y=upper_bound[country],
                                  name='+' + country[1], mode='lines',
                                  line_dash='dot',
                                  line_color=colors[i%n_colors],
                                  showlegend=False,
-                                 visible=False,
+                                 visible=True,
                                  hoverinfo='skip',
                                  line_width=.8))
         fig.add_trace(go.Scatter(x=lower_bound.index,
@@ -128,7 +132,7 @@ def make_timeplot(df_measure, df_prediction):
                                  line_dash='dot',
                                  line_color=colors[i%n_colors],
                                  showlegend=False,
-                                 visible=False,
+                                 visible=True,
                                  hoverinfo='skip',
                                  line_width=.8))