LTOM.01.002 Data Analysis in Community Ecology
Carlos P. Carmona (carlos.perez.carmona@ut.ee)
Aurele Toussaint (aurele.toussaint@ut.ee)
Meelis Pärtel (meelis.partel@ut.ee)
Diego P. F. Trindade (diego.trindade@ut.ee)
Slendy Rodriguez-Alarcon (slendy.rodriguez@ut.ee)
Autumn 2023
4 Scripts and data in R
4.1 Working directory
In which folder (working directory) you have your files?
wd=getwd() # get working directory
wd
## [1] "C:/Users/Carlos Perez Carmona/Dropbox/Courses/DataAnalysisCommunityEcology/2023-2024/Data"
setwd("C:/Users/Carlos Perez Carmona/Dropbox/Courses/DataAnalysisCommunityEcology/2023-2024/Data") # change to your favorite folder!
dir() # lists files
## [1] "basal_area.R" "bry.tab.txt" "climate.txt"
## [4] "clusters.rda" "community.rda" "coordinates.txt"
## [7] "Data.Rproj" "droughtnet.txt" "eesti reljeef.DAT"
## [10] "eesti reljeef.ID"
## [ reached getOption("max.print") -- omitted 26 entries ]Sometimes it is needed to give double backslashes:
"c:\\mp\\doc\\"
4.2 Keeping track of objects in the memory, saving R objects
You can check what objects you currently have and remove unnecessary.
# Making some objects
a=3
diversity=c(18,27,10,22,25,8,12)
soil.ph=c(4.5,6.4,3.9,5.3,5.9,3.2,4.2)
soil.type=c("mineral","mineral","peat","mineral","mineral","peat","peat")
ls()
## [1] "a" "diversity" "rootDir" "soil.ph" "soil.type" "wd"
ls.str() # more information
## a : num 3
## diversity : num [1:7] 18 27 10 22 25 8 12
## rootDir : chr "C:/Users/Carlos Perez Carmona/Dropbox/Courses/DataAnalysisCommunityEcology/2023-2024/Data"
## soil.ph : num [1:7] 4.5 6.4 3.9 5.3 5.9 3.2 4.2
## soil.type : chr [1:7] "mineral" "mineral" "peat" "mineral" "mineral" "peat" "peat"
## wd : chr "C:/Users/Carlos Perez Carmona/Dropbox/Courses/DataAnalysisCommunityEcology/2023-2024/Data"
save(diversity,soil.ph,soil.type,file="my.data.Rda") # saving so tahat R can use later
## NB! Other software cannot read these files!
rm(diversity,soil.ph,soil.type) # removes objects not needed
ls()
## [1] "a" "rootDir" "wd"
load("my.data.Rda") # loads what has been saved in R
ls()
## [1] "a" "diversity" "rootDir" "soil.ph" "soil.type" "wd"Explore with
?rmhow to remove all objects in the memory. WARNING, use carefully!
Answer
# currently we can still use this. Sometimes it is good to start your session with that expression
rm(list=ls())
# We now need to load our saved data again:
load("my.data.Rda")4.3 Creating own scripts
Scripts are just text files with codes. In RStudio you can add new R
Script file. When saving, use extension .R, for example
my_first_script.R
Add comments for your own, and for others with #
mark
Run your script by highlighting a part and then Cntr+Enter (=run!)
## My script, autumn 2019
# a is a variable ....
a = 5
print("Hello! This is my first script")
## [1] "Hello! This is my first script"
## end of my scriptSave the text above to a text file with name “my_first_script.R”.
Now we can call the same script.
a=NA
a
## [1] NA
source("my_first_script.R") # runs your script from the file
## Warning in file(filename, "r", encoding = encoding): cannot open file
## 'my_first_script.R': No such file or directory
## Error in file(filename, "r", encoding = encoding): cannot open the connection
a # a has been defined in the script!
## [1] NA4.4 Conditions in scripts
Does something only if conditions are fulfilled. The proportion of
script for which the condition applies can be given between special
parentheses { and }
soil="peat" # try to change to "mineral"
if (soil=="peat") { # commands for conditions are between {}
print("Has been wet")
}
## [1] "Has been wet"
answer=ifelse(soil.type=="peat","wet","dry") # conditions, if true, if not
print(answer)
## [1] "dry" "dry" "wet" "dry" "dry" "wet" "wet"Use function
ifelsewithplotto have scatterplot betweensoil.phanddiversityand select symbol colors depending onsoil.type
Answer
plot(soil.ph,diversity,col=ifelse(soil.type=="peat","brown","black"))
4.5 Loops in scripts
Loops will make a similar thing n times, for example with each
element of a vector. The part of script which is repeated should be
marked with { and }. Actually in R often you
can avoid loops since you can have logical questions to whole
vectors.
for (i in 1:7) { # i is index the loop is between {}
print(soil.type[i]) # within loop you need to add print to get output
}
## [1] "mineral"
## [1] "mineral"
## [1] "peat"
## [1] "mineral"
## [1] "mineral"
## [1] "peat"
## [1] "peat"
for (i in soil.type) { # now i is an element of the vector
print(i)
}
## [1] "mineral"
## [1] "mineral"
## [1] "peat"
## [1] "mineral"
## [1] "mineral"
## [1] "peat"
## [1] "peat"
soils.high.div=NULL # NULL is empty object, but we can define it!
for (i in 1:length(diversity)) {
if (diversity[i]>20) soils.high.div=c(soils.high.div,soil.type[i])
}
soils.high.div
## [1] "mineral" "mineral" "mineral"Please make
soils.high.divwithout loop.
Answer
soils.high.div=soil.type[diversity>20] # if possible, avoid loops!4.6 Data frames
We can have a single table which can include different types of data (e.g. numeric and character). We can also specify a grouping variable factor. When createing a data.frame, most non-numerical components are defined as factors. You can access individual columns of data.frame by name or by number or defining with data_frame_name$column_name
my.data=data.frame(soil.ph,soil.type,diversity) # creating data frame
my.data
## soil.ph soil.type diversity
## 1 4.5 mineral 18
## 2 6.4 mineral 27
## 3 3.9 peat 10
## 4 5.3 mineral 22
## 5 5.9 mineral 25
## 6 3.2 peat 8
## 7 4.2 peat 12
str(my.data) # Note that in data-frame texts are transformed to factors with numeric levels!
## 'data.frame': 7 obs. of 3 variables:
## $ soil.ph : num 4.5 6.4 3.9 5.3 5.9 3.2 4.2
## $ soil.type: chr "mineral" "mineral" "peat" "mineral" ...
## $ diversity: num 18 27 10 22 25 8 12
my.data[,2] # accessing by rows and columns
## [1] "mineral" "mineral" "peat" "mineral" "mineral" "peat" "peat"
names(my.data)
## [1] "soil.ph" "soil.type" "diversity"
colnames(my.data) # same!
## [1] "soil.ph" "soil.type" "diversity"
my.data[,"soil.type"] # accessing a column by name
## [1] "mineral" "mineral" "peat" "mineral" "mineral" "peat" "peat"
my.data$soil.type # accessing by $ symbol
## [1] "mineral" "mineral" "peat" "mineral" "mineral" "peat" "peat"
as.character(my.data$soil.type) # transforming factors to characters
## [1] "mineral" "mineral" "peat" "mineral" "mineral" "peat" "peat"
as.numeric(my.data$soil.type) # to text: factor levels!
## Warning: NAs introduced by coercion
## [1] NA NA NA NA NA NA NA
my.data$high.diversity=my.data$diversity>20 # adds a new logical column
my.data
## soil.ph soil.type diversity high.diversity
## 1 4.5 mineral 18 FALSE
## 2 6.4 mineral 27 TRUE
## 3 3.9 peat 10 FALSE
## 4 5.3 mineral 22 TRUE
## 5 5.9 mineral 25 TRUE
## 6 3.2 peat 8 FALSE
## 7 4.2 peat 12 FALSE4.7 Reading and exploring data from text file
We generally read data from files. We can read text files by setting column limits (space, comma, semicolon etc.), defining headers (column titles), column for row names etc. etc.
Save file ´envir.txt´to your R working directory!
envir=read.table("envir.txt",header=T, sep=" ",row.names = 1)
# Let's explore the file!
ncol(envir) # number of columns
## [1] 12
nrow(envir) # ... rows
## [1] 30
dim(envir) # dimensions
## [1] 30 12
names(envir) # names of columns (colnames works as well)
## [1] "lon" "lat" "forest.types" "Jrk..nr." "Proovi.nimi"
## [6] "pH.KCl" "N.." "P.mg.kg" "K.mg.kg" "Ca.mg.kg"
## [11] "Mg.mg.kg" "OA.."
row.names(envir) # if there are row names
## [1] "1" "2" "3" "4" "5" "6" "7" "8" "9" "10" "11" "12" "13" "14" "15"
## [16] "16" "17" "18" "19" "20" "21" "26" "27" "28" "31" "32" "33" "36" "44" "73"
str(envir)
## 'data.frame': 30 obs. of 12 variables:
## $ lon : num 26.5 26.4 26.4 26.3 26.2 ...
## $ lat : num 58.2 58.2 58.2 58.2 58.2 ...
## $ forest.types: int 1152 1131 1153 1512 1132 1132 1161 1162 1132 1161 ...
## $ Jrk..nr. : int 1 2 3 4 5 6 7 8 9 10 ...
## $ Proovi.nimi : chr "Vapramae" "Illi" "Vitipalu" "Konguta" ...
## $ pH.KCl : num 3.72 3.65 4.25 4.4 2.62 3.28 5.77 4.01 2.51 4.04 ...
## $ N.. : num 0.425 0.136 0.25 2.319 0.37 ...
## $ P.mg.kg : num 31.6 22.3 58 63 25.6 ...
## $ K.mg.kg : num 79 42.7 60.7 131.2 131.6 ...
## $ Ca.mg.kg : num 370 154 649 11208 282 ...
## $ Mg.mg.kg : num 43.9 21.6 83.3 601.5 74.6 ...
## $ OA.. : num 11.4 5.24 7.78 70.85 24.83 ...
head(envir) # just upper rows
## lon lat forest.types Jrk..nr. Proovi.nimi pH.KCl N.. P.mg.kg
## 1 26.45699 58.24387 1152 1 Vapramae 3.72 0.4246261 31.57133
## 2 26.43730 58.19846 1131 2 Illi 3.65 0.1364021 22.26459
## 3 26.41474 58.15658 1153 3 Vitipalu 4.25 0.2500752 58.04788
## 4 26.27875 58.21060 1512 4 Konguta 4.40 2.3192906 63.00445
## 5 26.15339 58.22432 1132 5 Vehendi 2.62 0.3698321 25.62339
## 6 26.32119 58.28426 1132 6 Erumae 3.28 0.2426881 16.21594
## K.mg.kg Ca.mg.kg Mg.mg.kg OA..
## 1 79.04454 370.4051 43.87278 11.397929
## 2 42.73691 154.3332 21.64893 5.237742
## 3 60.71264 648.9929 83.34118 7.777811
## 4 131.18302 11208.0226 601.45503 70.845348
## 5 131.56542 281.8415 74.61169 24.833289
## 6 66.16427 260.2084 25.95761 7.988277
summary(envir) # short summary of each variable
## lon lat forest.types Jrk..nr.
## Min. :26.11 Min. :58.09 Min. :1131 Min. : 1.00
## 1st Qu.:26.44 1st Qu.:58.17 1st Qu.:1132 1st Qu.: 8.25
## Median :26.78 Median :58.24 Median :1147 Median :15.50
## Mean :26.77 Mean :58.22 Mean :1194 Mean :18.70
## 3rd Qu.:27.02 3rd Qu.:58.27 3rd Qu.:1161 3rd Qu.:26.75
## Max. :27.42 Max. :58.36 Max. :1512 Max. :73.00
## Proovi.nimi pH.KCl N.. P.mg.kg
## Length:30 Min. :2.510 Min. :0.1189 Min. : 8.215
## Class :character 1st Qu.:2.965 1st Qu.:0.2207 1st Qu.:12.889
## Mode :character Median :3.525 Median :0.2575 Median :17.712
## Mean :3.593 Mean :0.3812 Mean :25.488
## 3rd Qu.:4.010 3rd Qu.:0.3751 3rd Qu.:31.160
## Max. :5.770 Max. :2.3193 Max. :65.774
## K.mg.kg Ca.mg.kg Mg.mg.kg OA..
## Min. : 32.93 Min. : 80.24 Min. : 18.71 Min. : 4.871
## 1st Qu.: 66.69 1st Qu.: 225.05 1st Qu.: 43.89 1st Qu.: 7.045
## Median : 80.91 Median : 430.02 Median : 76.38 Median : 8.408
## Mean : 87.97 Mean : 817.52 Mean :101.06 Mean :12.627
## 3rd Qu.: 98.67 3rd Qu.: 617.13 3rd Qu.:105.18 3rd Qu.:11.693
## Max. :167.19 Max. :11208.02 Max. :601.46 Max. :70.845
envir[1:4,6:12] # short subset
## pH.KCl N.. P.mg.kg K.mg.kg Ca.mg.kg Mg.mg.kg OA..
## 1 3.72 0.4246261 31.57133 79.04454 370.4051 43.87278 11.397929
## 2 3.65 0.1364021 22.26459 42.73691 154.3332 21.64893 5.237742
## 3 4.25 0.2500752 58.04788 60.71264 648.9929 83.34118 7.777811
## 4 4.40 2.3192906 63.00445 131.18302 11208.0226 601.45503 70.845348
my.cols=c(6:9,5) # pH, N, P, K and site name
soil.data=envir[,my.cols] # subset by selected columns to new data.frame
head(envir[,-c(1:2,4,6:12)]) ## leaving out some columns (not "-" before "c")
## forest.types Proovi.nimi
## 1 1152 Vapramae
## 2 1131 Illi
## 3 1153 Vitipalu
## 4 1512 Konguta
## 5 1132 Vehendi
## 6 1132 Erumae
my.rows=envir$pH.KCl>median(envir$pH.KCl) # logical vector
my.rows
## [1] TRUE TRUE TRUE TRUE FALSE FALSE TRUE TRUE FALSE TRUE FALSE FALSE
## [13] TRUE FALSE TRUE FALSE FALSE TRUE FALSE TRUE FALSE TRUE FALSE FALSE
## [25] TRUE FALSE TRUE FALSE FALSE TRUE
high.ph.soils=soil.data[my.rows,]
low.ph.soils=soil.data[!my.rows,] # note !
high.ph.soils.1=subset(soil.data, pH.KCl>median(pH.KCl)) # subset function
high.ph.soils==high.ph.soils.1 # comparing all values individually
## pH.KCl N.. P.mg.kg K.mg.kg Proovi.nimi
## 1 TRUE TRUE TRUE TRUE TRUE
## 2 TRUE TRUE TRUE TRUE TRUE
## 3 TRUE TRUE TRUE TRUE TRUE
## 4 TRUE TRUE TRUE TRUE TRUE
## 7 TRUE TRUE TRUE TRUE TRUE
## 8 TRUE TRUE TRUE TRUE TRUE
## 10 TRUE TRUE TRUE TRUE TRUE
## 13 TRUE TRUE TRUE TRUE TRUE
## 15 TRUE TRUE TRUE TRUE TRUE
## 18 TRUE TRUE TRUE TRUE TRUE
## 20 TRUE TRUE TRUE TRUE TRUE
## 26 TRUE TRUE TRUE TRUE TRUE
## 31 TRUE TRUE TRUE TRUE TRUE
## 33 TRUE TRUE TRUE TRUE TRUE
## 73 TRUE TRUE TRUE TRUE TRUE
all.equal(high.ph.soils,high.ph.soils.1) # comparing two objects
## [1] TRUE
sample(nrow(soil.data)) # random order
## [1] 25 12 26 22 11 3 21 23 18 29 6 30 1 28 14 9 20 13 5 7 19 17 10 24 15
## [26] 16 27 4 2 8
rand.soils=soil.data[sample(nrow(soil.data)),]make a separate data.frame xy for coordinates lon and lat and site name.
Answer
xy=envir[,c(1,2,5)]4.8 Writing data tables to text files
write.table(xy,file="coordinates.txt") # default settings
write.table(soil.data,file="soil.data.csv", sep=";", row.names = F) # own settings4.9 R packages
R has some basic packages but you need to install packages for special tasks. Let’s install package “readxl” which allows to read excel files.
# install.packages("readxl") # if you do not have this, please omit # and install
library(readxl) # even if installed, you should call the package library each R session with the function library
test=read_excel("excel.xlsx")
test # a bit strange format
## # A tibble: 7 × 3
## soil.ph soil.type diversity
## <dbl> <chr> <dbl>
## 1 4.5 mineral 18
## 2 6.4 mineral 27
## 3 3.9 peat 10
## 4 5.3 mineral 22
## 5 5.9 mineral 25
## 6 3.2 peat 8
## 7 4.2 peat 12
test=as.data.frame(test) # making a siple data.frame
test
## soil.ph soil.type diversity
## 1 4.5 mineral 18
## 2 6.4 mineral 27
## 3 3.9 peat 10
## 4 5.3 mineral 22
## 5 5.9 mineral 25
## 6 3.2 peat 8
## 7 4.2 peat 12