10 Pre-processing of ionomic data

Code

# pkg
library(readxl)
library(tidyverse)
library(dplyr)
 
# src
# source(here::here("src/function/stat_function/stat_analysis_main.R")) # for make plot 
# source("src/function/graph_function.R")

10.1 Data importation

Code

plant_info=read_excel(here::here("data/plant_information.xlsx"))

Code

# Creation of one function who analyse eatch excel for root, leaf and steam
source(here::here("src/function/iono/import_data_ionomic.R"))

#data importation ----
list_name_excel=c("ICP-2022-04E-M. Prudent_modif.xlsx","ICP-2022-06E-M. Prudent_modif.xlsx","ICP-2022-07E-M. Prudent_modif.xlsx")
list_range=c("A42:AE95","A39:AE92","A40:AE93")
list_range_colname=c("B25:AE25","B25:AE25","B25:AE25")
list_compartiment=c("leaf","root","stem")

df_ionomic=data.frame()
for (i in 1:length(list_name_excel)){
  print(list_compartiment[i])
  df_ionomic_x=import_data_ionomic(name_excel_x=list_name_excel[i],range_data_x=list_range[i],colname_x=list_range_colname[i],compartiment_x=list_compartiment[i])
  df_ionomic=rbind(df_ionomic,df_ionomic_x)
}

df_ionomic$only_C_and_N=NA

# Importation of CHN analyse 
df_chn=read_excel(here::here("data/iono/CHN/tableau résultats Corentin_compile.xlsx"))
df_chn_vertical=df_chn %>% 
  pivot_longer(c("N","C"),values_to="concentration",names_to="compound")
df_chn_vertical=df_chn_vertical %>% 
  #as_tibble(df_chn_vertical) %>% 
  dplyr::rename(id=Echantillon) %>% 
  as.data.frame()

df_chn_vertical=df_chn_vertical %>%
  group_by(id,compound) %>%
  dplyr::summarise(concentration=mean(concentration))

df_chn_vertical$concentration=df_chn_vertical$concentration*10000

df_chn_vertical$compartiment_code=substr(df_chn_vertical$id,start=1,stop=1)
df_chn_vertical$plant_num=substr(df_chn_vertical$id,2,nchar(df_chn_vertical$id))

df_chn_vertical$compartiment[df_chn_vertical$compartiment_code=="L"] <- "leaf"
df_chn_vertical$compartiment[df_chn_vertical$compartiment_code=="S"] <- "stem"
df_chn_vertical$compartiment[df_chn_vertical$compartiment_code=="R"] <- "root"


df_chn_vertical$compartiment_compound=paste0(df_chn_vertical$compartiment, "_", df_chn_vertical$compound)
df_chn_vertical$compound_compartiment=paste0(df_chn_vertical$compound, "_", df_chn_vertical$compartiment)
df_chn_vertical$justesse=1
df_chn_vertical$color="vert"

plant_info=read_excel(here::here("data/plant_information.xlsx"))
df_chn_vertical=merge(plant_info,df_chn_vertical,by="plant_num",all.y=T)

# Add column only_C_N on plant_info 
# df_chn_vertical$only_C_and_N=T

# Convert x in only_C_N into  into T
df_chn_vertical$only_C_and_N= ifelse(df_chn_vertical$only_C_and_N=="x",T,NA)

df_ionomic=rbind(df_ionomic,df_chn_vertical)

# Allocation and quantity of ion in eatch plant
#1- Importation of the data and merge
BM=read.csv2(here::here("data/physio/global_physio.csv"))
df_ionomic$plant_num=as.integer(df_ionomic$plant_num)
df_ionomic=merge(df_ionomic, BM %>% dplyr::select(plant_num, weight_root,weight_leaf, weight_stem, leaf_area, stem_area), by= "plant_num", all=T) %>% drop_na(concentration)
df_ionomic$weight_total=df_ionomic$weight_leaf+df_ionomic$weight_root+df_ionomic$weight_stem

# Corection of the justesse with color
# Justesse
df_ionomic$note=NA
df_ionomic$correct_concentration=NA
for( i in 1:length(df_ionomic$plant_num)){
  if(is.na(df_ionomic$color[i])){next}
  else if(df_ionomic$color[i]=="jaune"){
    df_ionomic$correct_concentration[i]=df_ionomic$concentration[i]/df_ionomic$justesse[i]
  }else if(df_ionomic$color[i]=="vert"){
    df_ionomic$correct_concentration[i]=df_ionomic$concentration[i]
  }else if (df_ionomic$color[i]=="rouge"){
    df_ionomic$correct_concentration[i]=NA
    df_ionomic$note[i]="[justesse color is red]"
  }
}

test=df_ionomic %>% filter(is.na(concentration))

# By organe
df_ionomic$qty=NA
for(i in 1:length(df_ionomic$plant_num)){
  if(is.na(df_ionomic$compartiment_code[i])){next}
  else if(df_ionomic$compartiment_code[i]=="L"){
    df_ionomic$qty[i]=df_ionomic$correct_concentration[i]*df_ionomic$weight_leaf[i]
  }else if(df_ionomic$compartiment_code[i]=="S"){
    df_ionomic$qty[i]=df_ionomic$correct_concentration[i]*df_ionomic$weight_stem[i]
  }else if(df_ionomic$compartiment_code[i]=="R"){
    df_ionomic$qty[i]=df_ionomic$correct_concentration[i]*df_ionomic$weight_root[i]
  }
}

# Allocation in %
df_ionomic=df_ionomic %>% 
  group_by(plant_num, compound) %>%
  mutate(sum_by_component=sum(qty, na.rm = T),
         num_compartiment=n()) %>% 
  mutate(allocation=ifelse(num_compartiment==3,qty/sum_by_component,NA))

# Create id for ionomic
df_ionomic$id_sample=paste0( df_ionomic$plant_num, "_", df_ionomic$compartiment_code)
df_ionomic$condition_organe=paste0(df_ionomic$condition,"_",df_ionomic$compartiment)

# Add type ion
df_ionomic$type_ion=NA
df_ionomic$type_ion_num=NA
for (i in 1:length(df_ionomic$compound)){
  if(
      df_ionomic$compound[i]=="C" || 
      df_ionomic$compound[i]=="N" ||
      df_ionomic$compound[i]=="Mg24"||
      df_ionomic$compound[i]=="P31"||
      df_ionomic$compound[i]=="S32"||
      df_ionomic$compound[i]=="K39"||
      df_ionomic$compound[i]=="Ca44"
     ){
      df_ionomic$type_ion[i]="essential_macroelement"
      df_ionomic$type_ion_num[i]=1
  } else if (
      df_ionomic$compound[i]=="B11" || 
      df_ionomic$compound[i]=="Mn55"||
      df_ionomic$compound[i]=="Fe56"||
      df_ionomic$compound[i]=="Ni60"||
      df_ionomic$compound[i]=="Cu63"||
      df_ionomic$compound[i]=="Zn66"||
      df_ionomic$compound[i]=="Mo98"||
      df_ionomic$compound[i]=="Mo95"
  ){
    df_ionomic$type_ion[i]="essential_microelement"
    df_ionomic$type_ion_num[i]=2
  }
  else if (
    df_ionomic$compound[i]=="Na23" || 
    df_ionomic$compound[i]=="Si28"||
    df_ionomic$compound[i]=="Al27"||
    df_ionomic$compound[i]=="V51"||
    df_ionomic$compound[i]=="Co59"||
    df_ionomic$compound[i]=="Se77"
  ){
    df_ionomic$type_ion[i]="beneficial_element"
    df_ionomic$type_ion_num[i]=3
  }
  else if (
    df_ionomic$compound[i]=="Ag107" || 
    df_ionomic$compound[i]=="As75"||
    df_ionomic$compound[i]=="Ba137"||
    df_ionomic$compound[i]=="Be9"||
    df_ionomic$compound[i]=="Cd111"||
    df_ionomic$compound[i]=="Cr52"|| 
    df_ionomic$compound[i]=="Pb208"||
    df_ionomic$compound[i]=="Rb85"||
    df_ionomic$compound[i]=="Sb121"||
    df_ionomic$compound[i]=="Ti49"||
    df_ionomic$compound[i]=="Tl205"
  ){
    df_ionomic$type_ion[i]="other"
    df_ionomic$type_ion_num[i]=4
  }
}

10.2 Filter outlier and cleaning data

10.2.1 Verif outlier

10.2.2 Extraction of good results only

Code

df_ionomic=df_ionomic %>%
  filter(!plant_num%in%c(1087,1039,93,1027)) %>% 
  filter(!compound%in%c("Al27","Si28","Ag107","Pb208","Mo98","Ti49","Sb121"))

I delet 93, 1027, 1039 and 1087

I not analyse Al, Si, Pb, Ag, Ti and S because the results are not interesting or the variability is too higher

Code

# Exportation of the results

df_ionomic_export<- apply(df_ionomic,2,as.character)
write.csv(as.data.frame(df_ionomic_export),row.names = F,here::here("data/iono/output/ionomic.csv"))
write.csv(as.data.frame(df_ionomic),row.names = F,here::here("data/iono/output/ionomic_with_na.csv"))

## Pre-processing of ionomic data {#sec-iono} ```{r setup,message=FALSE, warning=FALSE,echo=T} # pkg library(readxl) library(tidyverse) library(dplyr) # src # source(here::here("src/function/stat_function/stat_analysis_main.R")) # for make plot # source("src/function/graph_function.R") ``` ## Data importation ```{r} plant_info=read_excel(here::here("data/plant_information.xlsx")) ``` ```{r, results='hide'} # Creation of one function who analyse eatch excel for root, leaf and steam source(here::here("src/function/iono/import_data_ionomic.R")) #data importation ---- list_name_excel=c("ICP-2022-04E-M. Prudent_modif.xlsx","ICP-2022-06E-M. Prudent_modif.xlsx","ICP-2022-07E-M. Prudent_modif.xlsx") list_range=c("A42:AE95","A39:AE92","A40:AE93") list_range_colname=c("B25:AE25","B25:AE25","B25:AE25") list_compartiment=c("leaf","root","stem") df_ionomic=data.frame() for (i in 1:length(list_name_excel)){ print(list_compartiment[i]) df_ionomic_x=import_data_ionomic(name_excel_x=list_name_excel[i],range_data_x=list_range[i],colname_x=list_range_colname[i],compartiment_x=list_compartiment[i]) df_ionomic=rbind(df_ionomic,df_ionomic_x) } df_ionomic$only_C_and_N=NA # Importation of CHN analyse df_chn=read_excel(here::here("data/iono/CHN/tableau résultats Corentin_compile.xlsx")) df_chn_vertical=df_chn %>% pivot_longer(c("N","C"),values_to="concentration",names_to="compound") df_chn_vertical=df_chn_vertical %>% #as_tibble(df_chn_vertical) %>% dplyr::rename(id=Echantillon) %>% as.data.frame() df_chn_vertical=df_chn_vertical %>% group_by(id,compound) %>% dplyr::summarise(concentration=mean(concentration)) df_chn_vertical$concentration=df_chn_vertical$concentration*10000 df_chn_vertical$compartiment_code=substr(df_chn_vertical$id,start=1,stop=1) df_chn_vertical$plant_num=substr(df_chn_vertical$id,2,nchar(df_chn_vertical$id)) df_chn_vertical$compartiment[df_chn_vertical$compartiment_code=="L"] <- "leaf" df_chn_vertical$compartiment[df_chn_vertical$compartiment_code=="S"] <- "stem" df_chn_vertical$compartiment[df_chn_vertical$compartiment_code=="R"] <- "root" df_chn_vertical$compartiment_compound=paste0(df_chn_vertical$compartiment, "_", df_chn_vertical$compound) df_chn_vertical$compound_compartiment=paste0(df_chn_vertical$compound, "_", df_chn_vertical$compartiment) df_chn_vertical$justesse=1 df_chn_vertical$color="vert" plant_info=read_excel(here::here("data/plant_information.xlsx")) df_chn_vertical=merge(plant_info,df_chn_vertical,by="plant_num",all.y=T) # Add column only_C_N on plant_info # df_chn_vertical$only_C_and_N=T # Convert x in only_C_N into into T df_chn_vertical$only_C_and_N= ifelse(df_chn_vertical$only_C_and_N=="x",T,NA) df_ionomic=rbind(df_ionomic,df_chn_vertical) # Allocation and quantity of ion in eatch plant #1- Importation of the data and merge BM=read.csv2(here::here("data/physio/global_physio.csv")) df_ionomic$plant_num=as.integer(df_ionomic$plant_num) df_ionomic=merge(df_ionomic, BM %>% dplyr::select(plant_num, weight_root,weight_leaf, weight_stem, leaf_area, stem_area), by= "plant_num", all=T) %>% drop_na(concentration) df_ionomic$weight_total=df_ionomic$weight_leaf+df_ionomic$weight_root+df_ionomic$weight_stem # Corection of the justesse with color # Justesse df_ionomic$note=NA df_ionomic$correct_concentration=NA for( i in 1:length(df_ionomic$plant_num)){ if(is.na(df_ionomic$color[i])){next} else if(df_ionomic$color[i]=="jaune"){ df_ionomic$correct_concentration[i]=df_ionomic$concentration[i]/df_ionomic$justesse[i] }else if(df_ionomic$color[i]=="vert"){ df_ionomic$correct_concentration[i]=df_ionomic$concentration[i] }else if (df_ionomic$color[i]=="rouge"){ df_ionomic$correct_concentration[i]=NA df_ionomic$note[i]="[justesse color is red]" } } test=df_ionomic %>% filter(is.na(concentration)) # By organe df_ionomic$qty=NA for(i in 1:length(df_ionomic$plant_num)){ if(is.na(df_ionomic$compartiment_code[i])){next} else if(df_ionomic$compartiment_code[i]=="L"){ df_ionomic$qty[i]=df_ionomic$correct_concentration[i]*df_ionomic$weight_leaf[i] }else if(df_ionomic$compartiment_code[i]=="S"){ df_ionomic$qty[i]=df_ionomic$correct_concentration[i]*df_ionomic$weight_stem[i] }else if(df_ionomic$compartiment_code[i]=="R"){ df_ionomic$qty[i]=df_ionomic$correct_concentration[i]*df_ionomic$weight_root[i] } } # Allocation in % df_ionomic=df_ionomic %>% group_by(plant_num, compound) %>% mutate(sum_by_component=sum(qty, na.rm = T), num_compartiment=n()) %>% mutate(allocation=ifelse(num_compartiment==3,qty/sum_by_component,NA)) # Create id for ionomic df_ionomic$id_sample=paste0( df_ionomic$plant_num, "_", df_ionomic$compartiment_code) df_ionomic$condition_organe=paste0(df_ionomic$condition,"_",df_ionomic$compartiment) # Add type ion df_ionomic$type_ion=NA df_ionomic$type_ion_num=NA for (i in 1:length(df_ionomic$compound)){ if( df_ionomic$compound[i]=="C" || df_ionomic$compound[i]=="N" || df_ionomic$compound[i]=="Mg24"|| df_ionomic$compound[i]=="P31"|| df_ionomic$compound[i]=="S32"|| df_ionomic$compound[i]=="K39"|| df_ionomic$compound[i]=="Ca44" ){ df_ionomic$type_ion[i]="essential_macroelement" df_ionomic$type_ion_num[i]=1 } else if ( df_ionomic$compound[i]=="B11" || df_ionomic$compound[i]=="Mn55"|| df_ionomic$compound[i]=="Fe56"|| df_ionomic$compound[i]=="Ni60"|| df_ionomic$compound[i]=="Cu63"|| df_ionomic$compound[i]=="Zn66"|| df_ionomic$compound[i]=="Mo98"|| df_ionomic$compound[i]=="Mo95" ){ df_ionomic$type_ion[i]="essential_microelement" df_ionomic$type_ion_num[i]=2 } else if ( df_ionomic$compound[i]=="Na23" || df_ionomic$compound[i]=="Si28"|| df_ionomic$compound[i]=="Al27"|| df_ionomic$compound[i]=="V51"|| df_ionomic$compound[i]=="Co59"|| df_ionomic$compound[i]=="Se77" ){ df_ionomic$type_ion[i]="beneficial_element" df_ionomic$type_ion_num[i]=3 } else if ( df_ionomic$compound[i]=="Ag107" || df_ionomic$compound[i]=="As75"|| df_ionomic$compound[i]=="Ba137"|| df_ionomic$compound[i]=="Be9"|| df_ionomic$compound[i]=="Cd111"|| df_ionomic$compound[i]=="Cr52"|| df_ionomic$compound[i]=="Pb208"|| df_ionomic$compound[i]=="Rb85"|| df_ionomic$compound[i]=="Sb121"|| df_ionomic$compound[i]=="Ti49"|| df_ionomic$compound[i]=="Tl205" ){ df_ionomic$type_ion[i]="other" df_ionomic$type_ion_num[i]=4 } } ``` ## Filter outlier and cleaning data ### Verif outlier ### Extraction of good results only ```{r} df_ionomic=df_ionomic %>% filter(!plant_num%in%c(1087,1039,93,1027)) %>% filter(!compound%in%c("Al27","Si28","Ag107","Pb208","Mo98","Ti49","Sb121")) ``` ::: callout-important ## I delet 93, 1027, 1039 and 1087 ::: ::: callout-warning ## I not analyse Al, Si, Pb, Ag, Ti and S because the results are not interesting or the variability is too higher ::: ```{r} # Exportation of the results df_ionomic_export<- apply(df_ionomic,2,as.character) write.csv(as.data.frame(df_ionomic_export),row.names = F,here::here("data/iono/output/ionomic.csv")) write.csv(as.data.frame(df_ionomic),row.names = F,here::here("data/iono/output/ionomic_with_na.csv")) ```