Commit: a9e12042e2618e56ed24ce2f60c801058807a2b0 - myprojects-hg-reborn (hg)

Commit MetaInfo

• Revision: a9e12042e2618e56ed24ce2f60c801058807a2b0 (tree)
• Zeit: 2010-03-06 03:00:49
• Autor: lorenzo
• Commiter: lorenzo

Log Message

Some modifications to the code to analyze data and make the plots and also to the code which divides the interactions
into roles (the category ALL was also including categories which were anyway left out).

Ändern Zusammenfassung

• modified: Python-codes/compartimentize_contacts.py
• modified: R-codes/ggplot-boxplot-synoptics.R

Diff

diff -r fb267376c4a7 -r a9e12042e261 Python-codes/compartimentize_contacts.py
--- a/Python-codes/compartimentize_contacts.py	Thu Mar 04 17:19:26 2010 +0000
+++ b/Python-codes/compartimentize_contacts.py	Fri Mar 05 18:00:49 2010 +0000
@@ -75,7 +75,9 @@
 sel_P=s.where((tag_ids>=1400) & (tag_ids<1500))[0]
 sel_V=s.where((tag_ids>=1500) & (tag_ids<1600))[0]
 sel_A=s.where((tag_ids>=1600))[0]
-sel_ALL=s.where((tag_ids>=0))[0]
+#sel_ALL=s.where((tag_ids>=0))[0]
+sel_ALL=s.where((tag_ids<1100) | ((tag_ids>=1200) & (tag_ids<1300)) | ((tag_ids>=1300) & (tag_ids<1400)) \
+                | ((tag_ids>=1400) & (tag_ids<1500)) | ((tag_ids>=1600)) )[0]
 
 sel_list=[sel_E,sel_F,sel_N,sel_D,sel_P,sel_V,sel_A,sel_ALL]
 

diff -r fb267376c4a7 -r a9e12042e261 R-codes/ggplot-boxplot-synoptics.R
--- a/R-codes/ggplot-boxplot-synoptics.R	Thu Mar 04 17:19:26 2010 +0000
+++ b/R-codes/ggplot-boxplot-synoptics.R	Fri Mar 05 18:00:49 2010 +0000
@@ -254,6 +254,30 @@
 
 if (plot_before_network==1){
 
+
+file_loc <- c("../A-A/interactions_a_a_.dat",
+              "../D-D/interactions_d_d_.dat",
+              "../N-N/interactions_n_n_.dat",
+              "../P-P/interactions_p_p_.dat",
+              "../E-E/interactions_e_e_.dat")
+
+number_tags <- seq(length(file_loc))
+
+tag_categories <- c("A","D","N","P","E")
+
+
+for (i in seq(length(file_loc))){
+
+#fn <- paste(prefix_vec[i],"Contact Number", sep=" ")
+  
+tags_by_category <- read.table(file_loc[i])
+
+tags_by_category <- as.matrix(tags_by_category)
+
+number_tags[i] <- length(unique(c(tags_by_category[ ,2],tags_by_category[ ,3])))
+
+}
+  
 #This is useful and should get executed
 
 if (hist_plot==1){
@@ -287,16 +311,24 @@
 
 fn <- paste(prefix_vec[i],"Contact Number", sep=" ")
   
-my_dframe <- read.table(file_loc[i])
+my_dframe <- read.table(file_loc[i], header=FALSE)
 
 my_dframe <- as.matrix(my_dframe)
 
+if (i<=5){
+
+  print("dim(my_dframe)")
+  print(dim(my_dframe))
+
+}
+
 
 #i.e. I am using 1 hours as a max duration of each (uncumulated!!!) link.
 sel <- which(my_dframe<=filter_unc)
 
 my_dframe <- my_dframe[sel]
 
+
 ## print(max("my_dframe is "))
 
 ## print(max(my_dframe))
@@ -308,9 +340,12 @@
 names(my_dframe) <- c("time")
 #my_dframe$time <- factor(my_dframe$time)
 
-list_ggplot[[i]] <- ggplot(my_dframe,aes(x=time,y=..count..) )+
-  geom_histogram(binwidth=.2, colour="black", fill="blue",origin=log10(20))+
-  # geom_vline(xintercept = 3600)+
+list_ggplot[[i]] <- ggplot(my_dframe,aes(x=time) )+
+  geom_histogram(aes(y=..count..), binwidth=.2, colour="black", fill="blue",origin=log10(20))+
+    geom_vline(xintercept = log10(60), col="red", size=1)+
+#    geom_vline(xintercept = log10(600), col="black",size=1)+
+    geom_vline(xintercept = log10(600), col="darkgreen",size=1)+
+
   #xlab(theme_blank())+
   ylab(fn)+
 ##  scale_x_continuous(trans="log10", limits=c(19,3600))+
@@ -498,6 +533,22 @@
 
 
 
+prefix_vec_double <- c("A","A","A","D","A","N","A","P","A","E",
+                "D","D","D","N","D","P","D","E","N","N",
+                "N","P","E","N","P","P","E","P","E","E")
+
+
+
+norm_vec <- seq(length(prefix_vec))
+
+
+for (i in seq(length(prefix_vec))){
+  
+norm_vec[i] <- number_tags[which(tag_categories==prefix_vec_double[2*i-1])]*number_tags[which(tag_categories==prefix_vec_double[2*i])]
+
+}
+
+
 sum_of_contact_durations <- seq(length(prefix_vec))
 
 count_contacts <- seq(length(prefix_vec))
@@ -506,8 +557,8 @@
 
 for (i in seq(15)){
 
-print("i is, ")
-print(i)
+## print("i is, ")
+## print(i)
   
 fn <- paste(prefix_vec[i],"Contact Number", sep=" ")
   
@@ -552,9 +603,9 @@
 sum_of_contact_durations[i] <- sum(link_cumulative_durations)
 count_contacts[i] <- sum(number_contacts_in_link)
 
-if (prefix_vec[i]=="P-P"){
-pp_durations <- link_cumulative_durations
-}
+## if (prefix_vec[i]=="P-P"){
+## pp_durations <- link_cumulative_durations
+## }
 
 
 #Now I get rid of links whose cumulated duration is excessive
@@ -564,11 +615,11 @@
 
 my_dframe_tot <- cbind(link_cumulative_durations,number_contacts_in_link)
 
-print("max(my_dframe_tot[ ,1])")
-print(max(my_dframe_tot[ ,1]))
+## print("max(my_dframe_tot[ ,1])")
+## print(max(my_dframe_tot[ ,1]))
 
-print("max(my_dframe_tot[ ,2])")
-print(max(my_dframe_tot[ ,2]))
+## print("max(my_dframe_tot[ ,2])")
+## print(max(my_dframe_tot[ ,2]))
 
 
 my_dframe_tot <-  as.data.frame(my_dframe_tot)
@@ -671,7 +722,8 @@
 x_mid <- (x_max+x_min)/2
 
 y_min <- rep(0,length(prefix_vec))
-y_max <- sum_of_contact_durations/sum(sum_of_contact_durations)
+#y_max <- sum_of_contact_durations/sum(sum_of_contact_durations)
+y_max <- sum_of_contact_durations
 
 data_small <- cbind(y_max,y_min,x_min,x_max,x_mid)
 
@@ -684,14 +736,14 @@
 my_ybreaks <- (seq(8)-1)/10
 
 count_duration_plot<- ggplot(data_new)+
-  geom_rect(aes(ymin = y_min, ymax = y_max,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+  geom_rect(aes(ymin = y_min, ymax= y_max/3600.,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
   xlab("Role Pair")+
-  ylab("Fraction of Overall Contact Duration")+
+  ylab("Total Contact Duration [hours]")+
 ##  scale_x_continuous(trans="log10", limits=c(19,660))+
 ## scale_y_continuous(trans="log10", limits=c(1,700))+
   ## scale_x_continuous(trans="log10",limits=c(20,18000))+
   scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
-scale_y_continuous( limits=c(0.,0.7), breaks=my_ybreaks)+
+#scale_y_continuous( limits=c(0.,0.7), breaks=my_ybreaks)+
   opts(axis.title.x = theme_text(size = 35))+
   opts(axis.title.y = theme_text(size = 35, angle=90))+
  opts(axis.text.x = theme_text(size=30, colour="black",vjust=1))+
@@ -700,27 +752,27 @@
   opts(axis.text.y = theme_text(size=30, colour="black", hjust=1))
 
 
-CairoPDF("Contact_duration_fraction_per_coupled_roles_rectangles.pdf", width=20, height=10)
+CairoPDF("Contact_duration_per_coupled_roles.pdf", width=20, height=10)
 
 print(count_duration_plot)
 dev.off()
 
 
-#Now 
-
-data_new$y_max <- count_contacts/sum(count_contacts)
 
-my_ybreaks <- (seq(5)-1)/10
 
-count_contact_plot<- ggplot(data_new)+
-  geom_rect(aes(ymin = y_min, ymax = y_max,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+#and its normalized version
+
+my_ybreaks <- seq(0,20,by=5)
+
+count_duration_plot<- ggplot(data_new)+
+  geom_rect(aes(ymin = y_min, ymax = y_max/norm_vec/60.,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
   xlab("Role Pair")+
-  ylab("Fraction of Overall Contact Number")+
+  ylab("Normalized Total Contact Duration [min]")+
 ##  scale_x_continuous(trans="log10", limits=c(19,660))+
 ## scale_y_continuous(trans="log10", limits=c(1,700))+
   ## scale_x_continuous(trans="log10",limits=c(20,18000))+
   scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
-scale_y_continuous( limits=c(0.0,0.4), breaks=my_ybreaks)+
+scale_y_continuous( limits=c(0.,20.), breaks=my_ybreaks)+
   opts(axis.title.x = theme_text(size = 35))+
   opts(axis.title.y = theme_text(size = 35, angle=90))+
  opts(axis.text.x = theme_text(size=30, colour="black",vjust=1))+
@@ -729,11 +781,78 @@
   opts(axis.text.y = theme_text(size=30, colour="black", hjust=1))
 
 
-CairoPDF("Contact_number_fraction_per_coupled_roles_rectangles.pdf", width=20, height=10)
+CairoPDF("Normalized_contact_duration_per_coupled_roles.pdf", width=20, height=10)
+
+print(count_duration_plot)
+dev.off()
+
+
+
+
+
+#Now the fraction of contacts
+
+#data_new$y_max <- count_contacts/sum(count_contacts)
+
+data_new$y_max <- count_contacts
+
+
+my_ybreaks <- (seq(5)-1)/10
+
+count_contact_plot<- ggplot(data_new)+
+  geom_rect(aes(ymin = y_min, ymax = y_max,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+  xlab("Role Pair")+
+  ylab("Total Contact Number")+
+##  scale_x_continuous(trans="log10", limits=c(19,660))+
+## scale_y_continuous(trans="log10", limits=c(1,700))+
+  ## scale_x_continuous(trans="log10",limits=c(20,18000))+
+  scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
+#scale_y_continuous( limits=c(0.0,0.4), breaks=my_ybreaks)+
+  opts(axis.title.x = theme_text(size = 35))+
+  opts(axis.title.y = theme_text(size = 35, angle=90))+
+ opts(axis.text.x = theme_text(size=30, colour="black",vjust=1))+
+ # opts(axis.text.x = theme_blank())+
+#  geom_text(aes(label = labels, fill = NULL), data = data_new)+ 
+  opts(axis.text.y = theme_text(size=30, colour="black", hjust=1))
+
+
+CairoPDF("Contact_number_per_coupled_roles.pdf", width=20, height=10)
 
 print(count_contact_plot)
 dev.off()
 
+
+#and its normalized version
+
+my_ybreaks <- c(0,0.5e-3,1e-3,1.5e-3,2e-3)
+
+count_contact_plot<- ggplot(data_new)+
+  geom_rect(aes(ymin = y_min, ymax = y_max/norm_vec,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+  xlab("Role Pair")+
+  ylab("Normalized Total Contact Number")+
+##  scale_x_continuous(trans="log10", limits=c(19,660))+
+## scale_y_continuous(trans="log10", limits=c(1,700))+
+  ## scale_x_continuous(trans="log10",limits=c(20,18000))+
+  scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
+#scale_y_continuous( limits=c(0.0,2e-3), breaks=my_ybreaks)+
+  opts(axis.title.x = theme_text(size = 35))+
+  opts(axis.title.y = theme_text(size = 35, angle=90))+
+ opts(axis.text.x = theme_text(size=30, colour="black",vjust=1))+
+ # opts(axis.text.x = theme_blank())+
+#  geom_text(aes(label = labels, fill = NULL), data = data_new)+ 
+  opts(axis.text.y = theme_text(size=30, colour="black", hjust=1))
+
+
+CairoPDF("Normalized_contact_number_per_coupled_roles.pdf", width=20, height=10)
+
+print(count_contact_plot)
+dev.off()
+
+
+
+
+
+
 ##########################################################################################################à
 ##########################################################################################################à
 ##########################################################################################################à
@@ -777,8 +896,8 @@
 
 for (i in seq(length(prefix_vec))){
 
-print("i is, ")
-print(i)
+## print("i is, ")
+## print(i)
   
 #fn <- paste(prefix_vec[i]," Number of Contacts", sep=" ")
   
@@ -867,7 +986,8 @@
 x_mid <- (x_max+x_min)/2
 
 y_min <- rep(0,length(prefix_vec))
-y_max <- sum_of_contact_durations/sum(sum_of_contact_durations)
+#y_max <- sum_of_contact_durations/sum(sum_of_contact_durations)
+y_max <- sum_of_contact_durations
 
 data_small <- cbind(y_max,y_min,x_min,x_max,x_mid)
 
@@ -877,17 +997,17 @@
 
 names(data_new) <- c("y_max","y_min","x_min","x_max","x_mid","my_labels")
 
-my_ybreaks <- (seq(5)-1)/10
+my_ybreaks <- seq(0,200,by=50)
 
 count_duration_plot<- ggplot(data_new)+
-  geom_rect(aes(ymin = y_min, ymax = y_max,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+  geom_rect(aes(ymin = y_min, ymax = y_max/3600.,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
   xlab("Role")+
-  ylab("Fraction of Overall Contact Duration")+
+  ylab("Total Contact Duration [hours]")+
 ##  scale_x_continuous(trans="log10", limits=c(19,660))+
 ## scale_y_continuous(trans="log10", limits=c(1,700))+
   ## scale_x_continuous(trans="log10",limits=c(20,18000))+
   scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
-scale_y_continuous( limits=c(0.0,0.4), breaks=my_ybreaks)+
+scale_y_continuous( limits=c(0.0,200), breaks=my_ybreaks)+
   opts(axis.title.x = theme_text(size = 20))+
   opts(axis.title.y = theme_text(size = 20, angle=90))+
  opts(axis.text.x = theme_text(size=15, colour="black",vjust=1))+
@@ -896,27 +1016,25 @@
   opts(axis.text.y = theme_text(size=15, colour="black", hjust=1))
 
 
-CairoPDF("Contact_duration_fraction_per_role_rectangles.pdf")
+CairoPDF("Contact_duration_per_role.pdf")
 
 print(count_duration_plot)
 dev.off()
 
 
-#Now 
-
-data_new$y_max <- count_contacts/sum(count_contacts)
+#Now its normalized version 
 
-my_ybreaks <- (seq(4)-1)/10
+#my_ybreaks <- seq(0,200,by=50)
 
-count_contact_plot<- ggplot(data_new)+
-  geom_rect(aes(ymin = y_min, ymax = y_max,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+count_duration_plot<- ggplot(data_new)+
+  geom_rect(aes(ymin = y_min, ymax = y_max/number_tags/60.,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
   xlab("Role")+
-  ylab("Fraction of Overall Contact Number")+
+  ylab("Normalized Total Contact Duration [min]")+
 ##  scale_x_continuous(trans="log10", limits=c(19,660))+
 ## scale_y_continuous(trans="log10", limits=c(1,700))+
   ## scale_x_continuous(trans="log10",limits=c(20,18000))+
   scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
-scale_y_continuous( limits=c(0.0,0.3), breaks=my_ybreaks)+
+#scale_y_continuous( limits=c(0.0,20), breaks=my_ybreaks)+
   opts(axis.title.x = theme_text(size = 20))+
   opts(axis.title.y = theme_text(size = 20, angle=90))+
  opts(axis.text.x = theme_text(size=15, colour="black",vjust=1))+
@@ -925,11 +1043,71 @@
   opts(axis.text.y = theme_text(size=15, colour="black", hjust=1))
 
 
-CairoPDF("Contact_number_fraction_per_role_rectangles.pdf")
+CairoPDF("Normalized_contact_duration_per_role.pdf")
+
+print(count_duration_plot)
+dev.off()
+
+
+
+#####################################################################
+
+#Now the same quantity with contact numbers
+
+#data_new$y_max <- count_contacts/sum(count_contacts)
+data_new$y_max <- count_contacts
+
+my_ybreaks <- (seq(4)-1)/10
+
+count_contact_plot<- ggplot(data_new)+
+  geom_rect(aes(ymin = y_min, ymax = y_max,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+  xlab("Role")+
+  ylab("Total Contact Number")+
+##  scale_x_continuous(trans="log10", limits=c(19,660))+
+## scale_y_continuous(trans="log10", limits=c(1,700))+
+  ## scale_x_continuous(trans="log10",limits=c(20,18000))+
+  scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
+#scale_y_continuous( limits=c(0.0,0.3), breaks=my_ybreaks)+
+  opts(axis.title.x = theme_text(size = 20))+
+  opts(axis.title.y = theme_text(size = 20, angle=90))+
+ opts(axis.text.x = theme_text(size=15, colour="black",vjust=1))+
+ # opts(axis.text.x = theme_blank())+
+#  geom_text(aes(label = labels, fill = NULL), data = data_new)+ 
+  opts(axis.text.y = theme_text(size=15, colour="black", hjust=1))
+
+
+CairoPDF("Contact_number_per_role.pdf")
 
 print(count_contact_plot)
 dev.off()
 
+
+my_ybreaks <- c(0,0.5e-2,1e-2,1.5e-2,2e-2)
+
+count_contact_plot<- ggplot(data_new)+
+  geom_rect(aes(ymin = y_min, ymax = y_max/number_tags,xmin=x_min,xmax=x_max), colour="black", fill="blue")+
+  xlab("Role")+
+  ylab("Normalized Total Contact Number")+
+##  scale_x_continuous(trans="log10", limits=c(19,660))+
+## scale_y_continuous(trans="log10", limits=c(1,700))+
+  ## scale_x_continuous(trans="log10",limits=c(20,18000))+
+  scale_x_continuous(labels=prefix_vec, breaks=x_mid)+
+#scale_y_continuous( limits=c(0.0,0.02), breaks=my_ybreaks)+
+  opts(axis.title.x = theme_text(size = 20))+
+  opts(axis.title.y = theme_text(size = 20, angle=90))+
+ opts(axis.text.x = theme_text(size=15, colour="black",vjust=1))+
+ # opts(axis.text.x = theme_blank())+
+#  geom_text(aes(label = labels, fill = NULL), data = data_new)+ 
+  opts(axis.text.y = theme_text(size=15, colour="black", hjust=1))
+
+
+CairoPDF("Normalized_contact_number_per_role.pdf")
+
+print(count_contact_plot)
+dev.off()
+
+
+
 print("OK down here")
 
 #Now I go on with a boxplot
@@ -942,11 +1120,24 @@
 data_bp$role_label <- factor(data_bp$role_label)
 
 
+line_seq <- c(log10(300),log10(600),log10(1800))
+line_col <- c("5 minutes", "10 minutes", "30 minutes")
+
+data_line <- cbind(as.data.frame(line_seq),as.data.frame(line_col))
+
+names(data_line) <- c("line_seq","line_col")
 
 my_ybreaks <- c(1e1,1e2,1e3,1e4)
 gpl <- ggplot(data_bp,aes(x=role_label,y=contact_data) )+
  # geom_boxplot( quantiles = c(0,1))+
   geom_boxplot( )+
+     geom_hline(yintercept = log10(300), col="red", size=1)+
+    geom_hline(yintercept = log10(600), col="black",size=1)+
+     geom_hline(yintercept = log10(1800), col="darkgreen",size=1)+
+
+#geom_hline(yintercept = line_seq, col="red", size=1)+
+#geom_hline(data=dataline, yintercept = line_seq, color=line_col, size=1)+
+    
   ## scale_x_log10() +
   ## scale_y_log10()+  #better NOT to use short-hand notation if you want to set the axis
 #scale_x_continuous(trans = "log10", limits=c(10,1e4))+
@@ -1016,8 +1207,8 @@
 
 gt95 <- c(gt95,as.matrix(data_bp$contact_data[sel][sel2]))
 
-print ("length(gt95 is, )")
-print(length(gt95))
+## print ("length(gt95 is, )")
+## print(length(gt95))
 
 gt95_lab <- c(gt95_lab,as.vector(data_bp$role_label[sel][sel2]))
 
@@ -1045,6 +1236,10 @@
   geom_crossbar(aes(y = q50, ymin = q25, ymax = q75),fill="white") +
   geom_crossbar(aes(y = q95, ymin = q95, ymax = q95)) +
   geom_crossbar(aes(y = q5, ymin = q5, ymax = q5)) +
+    geom_hline(yintercept = log10(300), col="red", size=1)+
+    geom_hline(yintercept = log10(600), col="black",size=1)+
+    geom_hline(yintercept = log10(1800), col="darkgreen",size=1)+
+
 scale_y_continuous( trans="log10", limits=c(1e1,1e4), breaks=my_ybreaks)+
 
  geom_errorbar(limits) +
@@ -1091,8 +1286,9 @@
 
 if (plot_network==1){
 
+
 list_ggplot <- list() #I re-initialize this list
-
+#list_g_con <- list() #this  list will contain the connected nodes in the network
 
 ## file_loc <- c("../A-A/interactions_a_a_.dat",
 ##               "../A-D/interactions_a_d_.dat",
@@ -1185,6 +1381,24 @@
 
 
 
+file_loc4 <- c("../A-A/hashed_couples_in_contact_durations_1_.dat",
+              "../A-D/hashed_couples_in_contact_durations_1_.dat",
+              "../A-N/hashed_couples_in_contact_durations_1_.dat",
+              "../A-P/hashed_couples_in_contact_durations_1_.dat",
+              "../A-E/hashed_couples_in_contact_durations_1_.dat",
+              "../D-D/hashed_couples_in_contact_durations_1_.dat",
+              "../D-N/hashed_couples_in_contact_durations_1_.dat",
+              "../D-P/hashed_couples_in_contact_durations_1_.dat",
+              "../D-E/hashed_couples_in_contact_durations_1_.dat",
+              "../N-N/hashed_couples_in_contact_durations_1_.dat",
+              "../N-P/hashed_couples_in_contact_durations_1_.dat",
+              "../E-N/hashed_couples_in_contact_durations_1_.dat",
+              "../P-P/hashed_couples_in_contact_durations_1_.dat",
+              "../E-P/hashed_couples_in_contact_durations_1_.dat",
+              "../E-E/hashed_couples_in_contact_durations_1_.dat")
+
+
+
 
 prefix_vec <- c("A-A","A-D","A-N","A-P","A-E",
                 "D-D","D-N","D-P","D-E","N-N",
@@ -1206,8 +1420,8 @@
 #for (i in sel_num){
 
   
-print("i is, ")
-print(i)
+## print("i is, ")
+## print(i)
   
 
 duration_list <- read.table(file_loc2[i],header=FALSE)
@@ -1227,13 +1441,19 @@
 tab_ini <- read.table(file_loc[i],header=FALSE) 
 tab_ini <- as.matrix(tab_ini)
 
+hashed_ids <- read.table(file_loc4[i],header=FALSE)
+hashed_ids <- as.matrix(hashed_ids)
+
 
 #Now get rid of all the contacts below a certain threhshold
 
 tab_ini <- tab_ini[sel,]
 duration_list <- duration_list[sel]
 
+# and also update the duration list!
+duration_list <- duration_list[sel]
 
+hashed_ids <- hashed_ids[sel]
 
 ## graph_raw_no_time <- tab_ini[ , 2:dim(tab_ini)[2]]
 
@@ -1258,11 +1478,18 @@
 # 99878976  1234558
 #which are obviously the same edge repeated twice.
 
+#E(g)$weight <- duration_list
+
+#wb <- duration_list
 
 g <- simplify(g)
 
+#wa <- E(g)$weight
+
 names_g <- V(g)$name
 
+hashed_unique <- unique(hashed_ids)
+
 
 #E(g)$weight <- duration_list
 
@@ -1272,12 +1499,12 @@
 ids_isolated_nodes <- setdiff(tag_list,ids_connected_nodes)
 
 
-print("ids_isolated_nodes")
+## print("ids_isolated_nodes")
 
-print(ids_isolated_nodes)
+## print(ids_isolated_nodes)
 
-print("number isol nodes")
-print(length(ids_isolated_nodes))
+## print("number isol nodes")
+## print(length(ids_isolated_nodes))
 
 #Now add the isolated nodes to the graph
 
@@ -1380,4 +1607,8 @@
 dev.off()
 
 }
+
+
+
+
 print("So far so good")