Preparation for data analysis and data cleaning

Importing libraries

library(dplyr)
library(tidyverse)
library(base)
library(lubridate)
library(ggplot2)
library(webr)
library(plotly)

Importing data

data <- read.table(file = "C:\\Users\\Simona\\Desktop\\Personal_Portfolio\\Biobank_LT\\biobank.txt", header = TRUE, sep = ',')

data

Checking for the number of rows and columns in data set

glimpse(data)
## Rows: 17,750
## Columns: 15
## $ emimprefix   <chr> "TE", "TE", "TE", "TE", "TE", "TE", "TE", "TE", "TE", "TE…
## $ megnrsuffix  <chr> "N05", "N04", "N03", "N02", "N01", "A05", "A04", "A03", "…
## $ megmetai     <int> 2019, 2019, 2019, 2019, 2019, 2019, 2019, 2019, 2019, 201…
## $ megmen       <int> 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 12, 12, 12, 12, 12, 12, 12,…
## $ amzgr        <chr> "20-24", "20-24", "20-24", "20-24", "20-24", "20-24", "20…
## $ lytis        <chr> "V", "V", "V", "V", "V", "V", "V", "V", "V", "V", "V", "V…
## $ dgn          <chr> "", "", "", "", "", "", "", "", "", "", "C64", "C64", "C6…
## $ pt           <chr> "", "", "", "", "", "", "", "", "", "", "1b", "1b", "1b",…
## $ pn           <chr> "", "", "", "", "", "", "", "", "", "", "0", "0", "0", "0…
## $ pm           <chr> "", "", "", "", "", "", "", "", "", "", "0", "0", "0", "0…
## $ lokalizacija <chr> "Sėklidės", "Sėklidės", "Sėklidės", "Sėklidės", "Sėklidės…
## $ megpavad     <chr> "Navikas", "Navikas", "Navikas", "Navikas", "Navikas", "S…
## $ indpavad     <chr> "Kriomėgintuvėlis", "Kriomėgintuvėlis", "Kriomėgintuvėlis…
## $ kiekis       <dbl> 0.10, 0.10, 0.10, 0.32, 0.28, 0.11, 0.11, 0.11, 0.10, 0.1…
## $ vnt          <chr> "gr.", "gr.", "gr.", "gr.", "gr.", "gr.", "gr.", "gr.", "…

Merging the prefix and suffix to form ID

data$ID <- paste(data$emimprefix, data$megnrsuffix, sep = "")

data

Cheking if all samples have unique ID

unique(data$ID)
##   [1] "TEN05"   "TEN04"   "TEN03"   "TEN02"   "TEN01"   "TEA05"   "TEA04"  
##   [8] "TEA03"   "TEA02"   "TEA01"   "RENSL04" "RENSL03" "RENSL02" "RENSL01"
##  [15] "RENS02"  "RENS01"  "RENP04"  "RENP03"  "RENP02"  "RENP01"  "RENL02" 
##  [22] "RENL01"  "RENK04"  "RENK03"  "RENK02"  "RENK01"  "RENN05"  "RENN04" 
##  [29] "RENN03"  "RENN02"  "RENN01"  "RENA05"  "RENA04"  "RENA03"  "RENA02" 
##  [36] "RENA01"  "RENP05"  "RENSL05" "RENK07"  "RENK06"  "RENK05"  "RECSL04"
##  [43] "RECSL03" "RECSL02" "RECSL01" "RECS02"  "RECS01"  "RECP02"  "RECP01" 
##  [50] "RECL02"  "RECL01"  "RECK04"  "RECK03"  "RECK02"  "RECK01"  "RECP04" 
##  [57] "RECP03"  "RECN05"  "RECN04"  "RECN03"  "RECN02"  "RECN01"  "RECA05" 
##  [64] "RECA04"  "RECA03"  "RECA02"  "RECA01"  "PROSL04" "PROSL03" "PROSL02"
##  [71] "PROSL01" "PROS02"  "PROS01"  "PROP04"  "PROP03"  "PROP02"  "PROP01" 
##  [78] "PRON02"  "PRON01"  "PROL02"  "PROL01"  "PROK04"  "PROK03"  "PROK02" 
##  [85] "PROK01"  "PRON03"  "PROP05"  "PROSL05" "PROL03"  "PROS03"  "PLSL04" 
##  [92] "PLSL03"  "PLSL02"  "PLSL01"  "PLS02"   "PLS01"   "PLP04"   "PLP03"  
##  [99] "PLP02"   "PLP01"  
##  [ reached getOption("max.print") -- omitted 84 entries ]

Removing duplicates to get clean data set

df <- data %>% distinct()

df

Renaming columns from Lithuanian to English

df_renamed <-
df %>% 
  rename(age_group = amzgr,
    sex = lytis,
    diagnosis = dgn,
    localization = lokalizacija,
    stored_sample = megpavad,
    container = indpavad,
    amount = kiekis,
    measurmement = vnt,
    year = megmetai
  )

df_renamed

Changing localization name from Lithuanian to English

df_renamed_rv <-
df_renamed %>%
  mutate(localization = recode(localization,
                            "Gimdos kūnas" = "Uterus", 
                            "Inkstai" = "Kidneys",
                            "Kiaušidės" = "Ovaries", 
                            "Krūtis" = "Breast", 
                            "Melanoma" = "Melanoma", 
                            "Plaučiai" = "Lungs", 
                            "Prostatos biopsija" = "Prostate", 
                            "Sėklidės" ="Testicles", 
                            "Žarnynas" ="Intestine"))
df_renamed_rv

Removing not useful columns

df_full <- subset(df_renamed_rv, select = -c(megmen, emimprefix, megnrsuffix))

df_full

Changing diagnosis values to Cancerous and Non-Cancerous

df_full$diagnosis <- sub(".\\d+","Cancerous", df_full$diagnosis)

df_full
df_full$diagnosis <- sub("^$","Non-cancerous", df_full$diagnosis)

df_full

Visualization

Visualizing what are the most common localizations of stored samples by age group

breaks_y = c(0, 500, 1000, 1500, 2000, 2500, 3000, 3500, 4000)

plot1 <- ggplot(df_full, aes(x = age_group, fill = localization)) +
  geom_bar() +
  labs(x = "Age group", y = "Counts per Localization", title = "Stored Samples by Age Group at NVI Biobank", color = "Localization", fill = "Localization") +
  scale_y_continuous(limits = c(0, 4000), breaks = breaks_y) +
  theme_bw() +
  theme(axis.text = element_text(size = 9),
    axis.title.y = element_text(size = 12, face ="bold"),
    axis.title.x = element_text(size = 12, face ="bold"),
    legend.title=element_text(size=10, face ="bold"),
    legend.text=element_text(size=10),
    axis.text.x = element_text(angle = 45, hjust = 1),
    plot.title = element_text(size = 12, face = "bold"))

ggplotly(plot1)

Visualizing how stored sample count changed during the years 2018-2022 by sex and localization

breaks_y = c(0, 1000, 2000, 3000, 4000)

df_full_mod <- df_full %>%
  mutate(sex = recode(sex, "V" = "Male", "M" = "Female"))

plot2 <- ggplot(df_full_mod, aes(x = year, fill = localization)) +
  geom_bar() + 
  facet_wrap(vars(sex)) +
  labs(x = "Year of Collection", y = "Counts per Localization", title = "Stored Samples by Sex at NVI Biobank", color = "Localization", fill = "Localization") +
  scale_y_continuous(limits = c(0, 4000, breaks = breaks_y)) +
  scale_fill_discrete(labels=c("Gimdos kūnas" = "Uterus", 
     "Inkstai" = "Kidneys",
     "Kiaušidės" = "Ovaries", 
     "Krūtis" = "Breast", 
     "Melanoma" = "Melanoma", 
     "Plaučiai" = "Lungs", 
     "Prostatos biopsija" = "Prostate biopsy", 
     "Sėklidės" ="Testicles", 
     "Žarnynas" ="Intestine")) +
  theme_bw() +
  theme(axis.text = element_text(size = 9),
    axis.title.y = element_text(size = 12, face ="bold"),
    axis.title.x = element_text(size = 12, face ="bold"),
    legend.title=element_text(size=10, face ="bold"),
    legend.text=element_text(size=10),
    plot.title = element_text(size = 12, face = "bold")) 
ggplotly(plot2_flip)

Visualizing whether stored samples by localization and age group were cancerous or not

breaks_y = c(0, 500, 1000, 1500, 2000, 2500, 3000, 3500, 4000)

plot3 <- ggplot(df_full, aes(x = age_group, fill = localization)) +
  geom_bar()  +
  facet_wrap(vars(diagnosis)) +
  labs(x = "Age group", y = "Counts per Localization", title = "Stored Samples by Diagnosis and Age Group at NVI Biobank", color = "Localization", fill = "Localization") +
  scale_y_continuous(limits = c(0, 4000), breaks = breaks_y) +
  scale_fill_discrete(labels=c("Gimdos kūnas" = "Uterus", 
     "Inkstai" = "Kidneys",
     "Kiaušidės" = "Ovaries", 
     "Krūtis" = "Breast", 
     "Melanoma" = "Melanoma", 
     "Plaučiai" = "Lungs", 
     "Prostatos biopsija" = "Prostate biopsy", 
     "Sėklidės" ="Testicles", 
     "Žarnynas" ="Intestine")) +
  theme_bw() +
  theme(axis.text = element_text(size = 9),
    axis.title.y = element_text(size = 12, face ="bold"),
    axis.title.x = element_text(size = 12, face ="bold"),
    legend.title=element_text(size=10, face ="bold"),
    legend.text=element_text(size=10),
    axis.text.x = element_text(angle = 90, hjust = 1),
    plot.title = element_text(size = 12, face = "bold"))

ggplotly(plot3)