Dar roberd10, Update to v1

StudentData/comprehensive_LIBS.Rmd → AppRelated/comprehensive_LIBS.Rmd

@@ -28,10 +28,7 @@ if (!require("rmarkdown")) {
   install.packages("rmarkdown")
   library(rmarkdown)
 }
-if (!require("tidyverse")) {
-  install.packages("tidyverse")
-  library(tidyverse)
-}
+
 if (!require("stringr")) {
   install.packages("stringr")
   library(stringr)
@@ -96,6 +93,11 @@ if(!require("RColorBrewer")){
   install.packages("RColorBrewer")
   library(RColorBrewer)
 }
+
+if (!require("tidyverse")) {
+  install.packages("tidyverse")
+  library(tidyverse)
+}
 ```
 
 # Comprehensive LIBS Report
@@ -107,8 +109,8 @@ if(!require("RColorBrewer")){
 
 ```{r}
 # Raw, original LIBS dataset
-libs.df <- readRDS("/academics/MATP-4910-F24/DAR-Mars-F24/Data/supercam_libs_moc_loc.Rds")
-libs.df$point <- as.numeric(libs.df$point)
+libs.df <- readRDS("~/DAR-Mars-F24/StudentData/v1_libs.Rds")
+libs.df$Point <- as.numeric(libs.df$Point)
 
 # LIBS dataset w/o stdevs, other extra columns
 libs_trim.df <- libs.df %>% 
@@ -122,21 +124,22 @@ libs_earth.df <- readRDS("/academics/MATP-4910-F24/DAR-Mars-F24/Data/LIBS_traini
 Also included is PIXL, for some analyses
 ```{r}
 # PIXL data added
-pixl.df <- readRDS("/academics/MATP-4910-F24/DAR-Mars-F24/Data/samples_pixl_wide.Rds")
-pixl.df[sapply(pixl.df, is.character)] <- lapply(pixl.df[sapply(pixl.df, is.character)], 
-                                       as.factor)
+pixl.df <- readRDS("~/DAR-Mars-F24/StudentData/v1_pixl.Rds")
 pixl.df <- pixl.df[2:16,] #Excluding first, atmospheric sample
 # Make the matrix of just mineral percentage measurements
 pixl.matrix <- pixl.df[,2:14]
 pixl.matrix.scaled <- pixl.df[,2:14] %>% scale()
+# Sample meta data
+meta_full.df <- readRDS("~/DAR-Mars-F24/StudentData/v1_sample_meta.Rds")
+meta.df <- meta_full.df[2:16,]
 ```
 
 # LIBS Samples Plots
 
 ## Bray-Curtis Dissimilarity
 ```{r}
 # Matrix contain only the libs measurements for each mineral, Scaled
-libs.matrix.scaled <- as.matrix(libs.df[,6:13]) %>% scale(center=FALSE,scale=TRUE)
+libs.matrix.scaled <- as.matrix(libs.df[,7:22]) %>% scale(center=FALSE,scale=TRUE)
 
 # K-means clustering data w/o metadata, then adding targets back 
 km<-kmeans(libs.matrix.scaled,5)
@@ -191,7 +194,7 @@ ggtern(libs_ternary_clustered, ggtern::aes(x=x,y=y,z=z,cluster=cluster)) +
 
 ## LIBS Graphed by Location
 ```{r}
-libs_loc <- libs.df %>% select(c(SiO2, Al2O3, FeOT, MgO, CaO, Na2O, K2O, lat, lon))
+libs_loc <- libs.df %>% select(c(SiO2, Al2O3, FeOT, MgO, CaO, Na2O, K2O, Lat, Lon))
 libs_loc_ternary <- libs_loc %>%
   mutate(x=(SiO2+Al2O3)/100,y=(FeOT+MgO)/100,z=(CaO+Na2O+K2O)/100) %>%
   select(-c(SiO2,Al2O3,FeOT,MgO,CaO,Na2O,K2O)) %>%
@@ -204,7 +207,7 @@ tern.km <- kmeans(libs_ternary, k)
 libs_loc_ternary <- cbind(libs_loc_ternary, cluster=as.factor(tern.km$cluster))
 tern_clusters <- libs_ternary_clustered$cluster
 
-ggplot(libs_loc_ternary, aes(x=lon, y=lat, colour=cluster)) +
+ggplot(libs_loc_ternary, aes(x=Lon, y=Lat, colour=cluster)) +
     geom_point() +
     ggtitle("Clustered LIBS Data Graphed by Location")
 ```
@@ -213,10 +216,10 @@ ggplot(libs_loc_ternary, aes(x=lon, y=lat, colour=cluster)) +
 ```{r}
 suppressWarnings(
   libs.uniquetar <- 
-    aggregate(libs.df, list(Target = libs.df$target), mean))
+    aggregate(libs.df, list(Target = libs.df$Target), mean))
 
 #Aggregate data
-libs.uniquetar <- libs.uniquetar %>% select(!c(target,point))
+libs.uniquetar <- libs.uniquetar %>% select(!c(Target,Point))
 libs.uniquetar <- libs.uniquetar %>% 
   mutate(y = (FeOT + MgO) / 100, z = (CaO+Na2O+K2O) / 100, x = (SiO2 + Al2O3) / 100)
 libs_ternplot2 <- libs.uniquetar %>% select(c(x,y,z))
@@ -355,7 +358,7 @@ From the plot of the original data, we see that Cluster 1 (red) mostly falls in
 
 ## LIBS Samples Heatmap (clustered by k=5)
 ```{r}
-libs.matrix <- as.matrix(libs_trim.df[,6:13])
+libs.matrix <- as.matrix(libs_trim.df[,7:14])
 libs.scaled <- libs.matrix %>% scale()
 set.seed(500)
 k <- 5
@@ -366,8 +369,8 @@ pheatmap(km$centers,scale="none")
 
 ## LIBS PCA Plot
 ```{r}
-lib.matrix.scaled.pca <- prcomp(libs.scaled, scale=FALSE)
-ggbiplot::ggbiplot(lib.matrix.scaled.pca,
+libs.matrix.scaled.pca <- prcomp(libs.scaled, scale=FALSE)
+ggbiplot::ggbiplot(libs.matrix.scaled.pca,
 groups = as.factor(km$cluster)) +
 xlim(-5,2.5) + ylim(-2.5,7.5) +
 theme_bw()
@@ -399,7 +402,7 @@ ggtern(libs_target_ternary, ggtern::aes(x=x,y=y,z=z)) +
 
 ## Calibration Targets Violin Plot
 ```{r}
-libs_earth_trim <- libs.df %>% select(c(target, SiO2, TiO2, Al2O3, FeOT, MgO, CaO, Na2O, K2O, sol)) %>%
+libs_earth_trim <- libs.df %>% select(c(Target, SiO2, TiO2, Al2O3, FeOT, MgO, CaO, Na2O, K2O, Sol)) %>%
 rowwise() %>% mutate(Si= (SiO2-libs_earth.df[3,2])/(libs_earth.df[4,2] - libs_earth.df[2,2]),
 Ti= (TiO2-libs_earth.df[3,3])/(libs_earth.df[4,3] - libs_earth.df[2,3]),
 Al= (Al2O3-libs_earth.df[3,4])/(libs_earth.df[4,4] - libs_earth.df[2,4]),
@@ -408,7 +411,7 @@ Mg= (MgO-libs_earth.df[3,6])/(libs_earth.df[4,6] - libs_earth.df[2,6]),
 Ca= (CaO-libs_earth.df[3,7])/(libs_earth.df[4,7] - libs_earth.df[2,7]),
 Na= (Na2O-libs_earth.df[3,8])/(libs_earth.df[4,8] - libs_earth.df[2,8]),
 K= (K2O-libs_earth.df[3,9])/(libs_earth.df[4,9] - libs_earth.df[2,9])) %>%
-select(!c(SiO2, TiO2, Al2O3, FeOT, MgO, CaO, Na2O, K2O, target, sol))
+select(!c(SiO2, TiO2, Al2O3, FeOT, MgO, CaO, Na2O, K2O, Target, Sol))
 
 ggplot() +
 geom_violin(aes(y = libs_earth_trim$Si$SiO2, x = "Si"), fill = "blue") +
@@ -442,9 +445,10 @@ labs(title="Biplot of PCA on Earth Scaled LIBS")
 Clustered LIBS with PIXL samples clustered by campaign.
 ```{r}
 new_pixl_trim <- pixl.df %>%
-dplyr::select(c("Na20","Mgo","Al203","Si02", "K20","Cao","FeO-T", campaign, type)) %>%
-rename("Na2O"="Na20","MgO"="Mgo","Al2O3"="Al203","SiO2"="Si02","K2O"="K20",
-"CaO"="Cao","FeOT"="FeO-T")
+  dplyr::select(c("Na2O","MgO","Al2O3","SiO2", "K2O","CaO","FeOT"))
+Campaign <- meta.df$Campaign
+Type <- meta.df$Type
+new_pixl_trim <- cbind(new_pixl_trim,Campaign,Type)
 pixl_ternary <- new_pixl_trim %>%
   mutate(x=(SiO2+Al2O3)/100,y=(FeOT+MgO)/100,z=(CaO+Na2O+K2O)/100) %>%
   select(-c(SiO2,Al2O3,FeOT,MgO,CaO,Na2O,K2O)) %>%
@@ -465,11 +469,11 @@ z="Ca+Na2+K2") +
 #suppress warnings here because of some warning with aes()
 #add PIXL samples - atmospheric onto the ternary plot
 suppressWarnings(geom_point(
-data=pixl_ternary, ggtern::aes(x=x, y=y, z=z, cluster=campaign, shape=campaign),
+data=pixl_ternary, ggtern::aes(x=x, y=y, z=z, cluster=Campaign, shape=Campaign),
 size = 2)) +
 #Add labels to PIXL data corresponding to sample number
 suppressWarnings(geom_text(data=pixl_ternary,
-ggtern::aes(x=x, y=y, z=z, label=Sample_display, cluster=campaign,
+ggtern::aes(x=x, y=y, z=z, label=Sample_display, cluster=Campaign,
 hjust = ifelse(x > 0.43, 1, -0.1), # Horizontal adjust to avoid overlap
 vjust = ifelse(x == 0.3668, 1.3,
 ifelse(x == 0.375, 1, ifelse(x > 0.43, 1.5, -0.3))),
@@ -490,11 +494,11 @@ z="Ca+Na2+K2") +
 #suppress warnings here because of some warning with aes()
 #add PIXL samples - atmospheric onto the ternary plot
 suppressWarnings(geom_point(
-data=pixl_ternary, ggtern::aes(x=x, y=y, z=z, cluster=type, shape=type),
+data=pixl_ternary, ggtern::aes(x=x, y=y, z=z, cluster=Type, shape=Type),
 size = 2)) +
 #Add labels to PIXL data corresponding to sample number
 suppressWarnings(geom_text(data=pixl_ternary,
-ggtern::aes(x=x, y=y, z=z, label=Sample_display, cluster=type,
+ggtern::aes(x=x, y=y, z=z, label=Sample_display, cluster=Type,
 hjust = ifelse(x > 0.43, 1, -0.1), # Horizontal adjust to avoid overlap
 vjust = ifelse(x == 0.3668, 1.3,
 ifelse(x == 0.375, 1, ifelse(x > 0.43, 1.5, -0.3))),
@@ -508,10 +512,10 @@ Clustered LIBS, aggregated by target, with PIXL samples clustered by rock type
 #but thats is fine as we have the target anyways and point is no longer relevant
 suppressWarnings(
   libs.uniquetar <- 
-    aggregate(libs.df, list(Target = libs.df$target), mean))
+    aggregate(libs.df, list(Target = libs.df$Target), mean))
 
 #drop target and point from the data frame
-libs.uniquetar <- libs.uniquetar %>% select(!c(target,point))
+libs.uniquetar <- libs.uniquetar %>% select(!c(Target,Point))
 libs.uniquetar <- libs.uniquetar %>% 
   mutate(y = (FeOT + MgO) / 100, z = (CaO+Na2O+K2O) / 100, x = (SiO2 + Al2O3) / 100)
 libs_ternplot2 <- libs.uniquetar %>% select(c(x,y,z))
@@ -529,10 +533,10 @@ ggtern(libs_ternplot2, ggtern::aes(x=x, y=y, z=z,cluster=cluster)) +
        y="Fe+Mg",
        z="Ca+Na2+K2") +
   suppressWarnings(geom_point(
-    data=pixl_ternary, ggtern::aes(x=x, y=y, z=z, cluster=type, shape=type),
+    data=pixl_ternary, ggtern::aes(x=x, y=y, z=z, cluster=Type, shape=Type),
     size = 2.5)) +
   suppressWarnings(geom_text(data=pixl_ternary,
-            ggtern::aes(x=x, y=y, z=z, label=Sample_display, cluster=type,
+            ggtern::aes(x=x, y=y, z=z, label=Sample_display, cluster=Type,
             hjust = ifelse(x > 0.43, 1, -0.1),   # Horizontal adjust to avoid overlap
             vjust = ifelse(x == 0.3668, 1.3, 
                            ifelse(x == 0.375, 1, ifelse(x > 0.43, 1.5, -0.3))),
@@ -542,75 +546,80 @@ ggtern(libs_ternplot2, ggtern::aes(x=x, y=y, z=z,cluster=cluster)) +
 
 ## LIBS data plotted by Location, w/ PIXL
 ```{r}
-pixllibs <- readRDS("~/DAR-Mars-F24/StudentData/PIXL_LIBS_Combined.Rds")
-
-#make a filtered data frame that picks the max point out of all libs samples at a certain target 
-# for simplicity
-df_filtered <- pixllibs %>%
-  group_by(lat, lon) %>%
-  filter(point == max(point)) %>%
-  ungroup()
-df_filtered <- na.omit(df_filtered)
-df_filtered$lat <- as.numeric(df_filtered$lat)
-df_filtered$lon <- as.numeric(df_filtered$lon)
-
-#fix variable type and N/A issues in data frame
-pixllibs$lat <- as.numeric(pixllibs$lat)
-pixllibs$lon <- as.numeric(pixllibs$lon)
-pixllibs$Long <- as.numeric(pixllibs$Long)
-pixllibs_no_na <- na.omit(df_filtered)
-
-#make a data frame with the unique pixl coordinates since they are in pairs of identical lat/lon
-unique_pixl <- pixllibs_no_na %>%
-  select(lat, Long, name) %>% distinct()
-unique_pixl$Long <- as.numeric(unique_pixl$Long)
-
-#was using this to work on making a key for the plot, couldn't get it to work correctly.
-unique_pixl_with_label <- unique_pixl %>%
-  mutate(source = "PIXL")
-pixl_libs_with_label <- pixllibs %>%
-  mutate(source = "LIBS")
-
-#plot of libs and pixl data by lat/lon
-ggplot(data = pixllibs)  +
-  geom_point(mapping = aes(x = lat, y = lon), colour = 'blue') +
-  geom_point(mapping = aes(x = lat, y = Long), data = unique_pixl, colour = 'red') +
-  geom_text_repel(mapping = aes(x = lat, y = Long, label = name), data = unique_pixl, 
-            vjust = 2, colour = 'red') +
-  labs(title = "Libs and PIXL by Location",
-       x = "Latitude",
-       y = "Longitude",
-       color = "Point Type")
+# pixllibs <- readRDS("~/DAR-Mars-F24/StudentData/v1_libs_to_sample.Rds")
+# 
+# ## IMPORTANT: THIS AREA DOESN'T WORK BECAUSE MARGO HAS COMPLETELY SHIFTED WHAT THE FILE LOOKS LIKE SO THAT INSTEAD OF MENTIONING JUST WHICH IS CLOSEST WITHIN A RANGE, IT TELLS YOU WHAT THE DISTANCE BETWEEN EACH LIBS AND ITS CLOSEST PIXL. CHOOSE A THRESHOLD OF <= 7 AND REDO THIS AREA.
+# 
+# #make a filtered data frame that picks the max point out of all libs samples at a certain target
+# # for simplicity
+# df_filtered <- pixllibs %>%
+#   group_by(Lat, Lon) %>%
+#   filter(point == max(Point)) %>%
+#   ungroup()
+# df_filtered <- na.omit(df_filtered)
+# df_filtered$Lat <- as.numeric(df_filtered$Lat)
+# df_filtered$Lon <- as.numeric(df_filtered$Lon)
+# 
+# #fix variable type and N/A issues in data frame
+# pixllibs$lat <- as.numeric(pixllibs$Lat)
+# pixllibs$lon <- as.numeric(pixllibs$Lon)
+# pixllibs$Long <- as.numeric(pixllibs$Lon)
+# pixllibs_no_na <- na.omit(df_filtered)
+# 
+# #make a data frame with the unique pixl coordinates since they are in pairs of identical lat/lon
+# unique_pixl <- pixllibs_no_na %>%
+#   select(lat, Long, name) %>% distinct()
+# unique_pixl$Long <- as.numeric(unique_pixl$Long)
+# 
+# #was using this to work on making a key for the plot, couldn't get it to work correctly.
+# unique_pixl_with_label <- unique_pixl %>%
+#   mutate(source = "PIXL")
+# pixl_libs_with_label <- pixllibs %>%
+#   mutate(source = "LIBS")
+# 
+# #plot of libs and pixl data by lat/lon
+# ggplot(data = pixllibs)  +
+#   geom_point(mapping = aes(x = lat, y = lon), colour = 'blue') +
+#   geom_point(mapping = aes(x = lat, y = Long), data = unique_pixl, colour = 'red') +
+#   geom_text_repel(mapping = aes(x = lat, y = Long, label = name), data = unique_pixl,
+#             vjust = 2, colour = 'red') +
+#   labs(title = "Libs and PIXL by Location",
+#        x = "Latitude",
+#        y = "Longitude",
+#        color = "Point Type")
 ```
 
 ## LIBS and PIXL Sample Types Over Time
 ```{r}
-pixl.sol.coordinates<-readRDS("/academics/MATP-4910-F24/DAR-Mars-F24/StudentData/pixl_sol_coordinates.Rds")
-pixl.sol<-pixl.sol.coordinates[,c(3,18,22)]
+pixl.sol<-meta_full.df[1:16,c(3,2,6)]
 pixl.sol<-cbind(pixl.sol,"PIXL")
 
-libs.sol<-libs.df[,c(1,4,5)]
+libs.sol<-libs.df[,c(3,1,2)]
 
 libs.sol<- libs.sol %>%
-  group_by(sol, target) %>%
-  summarise(across(point,max))
+  group_by(Sol, Target) %>%
+  summarise(across(Point,max))
 
 libs.sol<-cbind(libs.sol,"type"="type")
 
 libs.sol<-libs.sol %>% 
-  mutate(type = ifelse(grepl("scct", target),"scct", type)) %>%
-  mutate(type= ifelse(grepl("scam", target) & type=="type","scam",type)) %>%
-  mutate(type= ifelse(grepl("aegis", target) & type=="type","aegis",type)) %>%
-  mutate(type=ifelse(grepl("[0-9]",target) & type=="type","num",type)) %>%
+  mutate(type= ifelse(grepl("scam", Target) & type=="type","scam",type)) %>%
+  mutate(type= ifelse(grepl("aegis", Target) & type=="type","aegis",type)) %>%
+  mutate(type=ifelse(grepl("[0-9]", Target) & type=="type","num",type)) %>%
   mutate(type=ifelse(type=="type","other",type))
 
+libs_scct.df <- readRDS("~/DAR-Mars-F24/StudentData/v1_libs_earth_references.Rds")
+libs_scct.sol <- cbind(libs_scct.df[,c(3,1,2)],"scct")
+
 colnames(libs.sol)<-c("Sol","Target_Name","Points_Abrasion","Type")
 colnames(pixl.sol)<-c("Sol","Target_Name","Points_Abrasion","Type")
+colnames(libs_scct.sol)<-c("Sol","Target_Name","Points_Abrasion","Type")
 
 pixl.sol$Sol<-as.numeric(pixl.sol$Sol)
 libs.sol$Points_Abrasion<-as.character(libs.sol$Points_Abrasion)
+libs_scct.sol$Points_Abrasion<-as.character(libs_scct.sol$Points_Abrasion)
 
-libsandpixl<-rbind(libs.sol,pixl.sol)
+libsandpixl<-rbind(libs.sol,libs_scct.sol,pixl.sol)
 libsandpixl$Type<-as.factor(libsandpixl$Type)
 
 ggplot()+
@@ -656,31 +665,34 @@ grid.arrange(plot1, plot2, plot3, plot4,plot5,plot6,ncol=1)
 
 ## LIBS Data according to its nearest PIXL cluster
 ```{r}
-libsandpixl<-readRDS("/academics/MATP-4910-F24/DAR-Mars-F24/StudentData/PIXL_LIBS_Combined.Rds")
-libsandpixl<-na.omit(libsandpixl)
-libsandpixl<-cbind("index"=0,libsandpixl)
-libsandpixl$index<-rownames(libsandpixl)
-
-libs_trim.df<-cbind("index"=0,libs_trim.df)
-libs_trim.df$index<-rownames(libs_trim.df)
-libs_trim.df<-libs_trim.df[libs_trim.df$index %in% libsandpixl$index,]
-
-libs.matrix <- as.matrix(libs_trim.df[,7:14]) 
-
-libs.tern <- as.data.frame(libs.matrix) %>%
-  mutate(x=(SiO2+Al2O3)/100,y=(FeOT+MgO)/100,z=(CaO+Na2O+K2O)/100) %>%
-  select(-c(SiO2,Al2O3,FeOT,MgO,CaO,Na2O,K2O,TiO2))
-
-libs.tern<-cbind(libs.tern,"Abrasion"=as.factor(libsandpixl$abrasion))
-
-ggtern(libs.tern, ggtern::aes(x=x,y=y,z=z)) +
-  geom_point(data=libs.tern,aes(color=Abrasion,alpha=0.5)) + 
-  theme_rgbw() + 
-  labs(title="Mars LIBS Corresponding to PIXL",
-       x="Si+Al",
-       y="Fe+Mg",
-       z="Ca+Na+K")+theme(legend.position="right") + 
-  geom_point(data=libs.tern,
-            aes(color=Abrasion,alpha=0.5)) +
-  guides(alpha="none")
+## IMPORTANT: Same problem as before, "PIXL_LIBS_Combined.Rds" has been changed and you should be using "v1_libs_to_sample.Rds". Additionally, I notice that you are calling this "libsandpixl" even though in the previous section you had a completely different data frame "libsandpixl" (containing completely different types of data). Also, this is the second time you are importing "PIXL_LIBS_Combined.Rds". If you get it working in the previous notes code block, you can probably just use that again here.
+
+
+# libsandpixl<-readRDS("/academics/MATP-4910-F24/DAR-Mars-F24/StudentData/PIXL_LIBS_Combined.Rds")
+# libsandpixl<-na.omit(libsandpixl)
+# libsandpixl<-cbind("index"=0,libsandpixl)
+# libsandpixl$index<-rownames(libsandpixl)
+# 
+# libs_trim.df<-cbind("index"=0,libs_trim.df)
+# libs_trim.df$index<-rownames(libs_trim.df)
+# libs_trim.df<-libs_trim.df[libs_trim.df$index %in% libsandpixl$index,]
+# 
+# libs.matrix <- as.matrix(libs_trim.df[,7:14]) 
+# 
+# libs.tern <- as.data.frame(libs.matrix) %>%
+#   mutate(x=(SiO2+Al2O3)/100,y=(FeOT+MgO)/100,z=(CaO+Na2O+K2O)/100) %>%
+#   select(-c(SiO2,Al2O3,FeOT,MgO,CaO,Na2O,K2O,TiO2))
+# 
+# libs.tern<-cbind(libs.tern,"Abrasion"=as.factor(libsandpixl$abrasion))
+# 
+# ggtern(libs.tern, ggtern::aes(x=x,y=y,z=z)) +
+#   geom_point(data=libs.tern,aes(color=Abrasion,alpha=0.5)) + 
+#   theme_rgbw() + 
+#   labs(title="Mars LIBS Corresponding to PIXL",
+#        x="Si+Al",
+#        y="Fe+Mg",
+#        z="Ca+Na+K")+theme(legend.position="right") + 
+#   geom_point(data=libs.tern,
+#             aes(color=Abrasion,alpha=0.5)) +
+#   guides(alpha="none")
 ```