In sbalci/histopathology-template: Template of R Codes Used in Histopathology Research.

Codes for clean and recode data.^[See childRmd/_09cleanRecode.Rmd file for other codes]

mydata <- janitor::clean_names(mydata)
# cat(names(mydata), sep = ",\n")
# names(mydata) <- c(names(mydata)[1:21], paste0("Soru", 1:30))

iris %>% 
  explore::clean_var(data = ., 
                     var = Sepal.Length,  
            min_val = 4.5, 
            max_val = 7.0, 
            na = 5.8, 
            name = "sepal_length") %>% 
  describe()

summarytools::view(summarytools::dfSummary(mydata))
dplyr::glimpse(mydata)

questionr::irec()

questionr::iorder()

questionr::icut()

iris %>% mutate(sumVar = rowSums(.[1:4]))

iris %>% mutate(sumVar = rowSums(select(., contains("Sepal")))) %>% head

iris %>% mutate(sumVar = select(., contains("Sepal")) %>% rowSums()) %>% head

library(finalfit)
# https://www.datasurg.net/2019/10/15/jama-retraction-after-miscoding-new-finalfit-function-to-check-recoding/
# intentionally miscoded
colon_s %>%
  mutate(
    sex.factor2 = forcats::fct_recode(sex.factor,
      "F" = "Male",
      "M" = "Female")
  ) %>%
  count(sex.factor, sex.factor2)

# Install
# devtools::install_github('ewenharrison/finalfit')
library(finalfit)
library(dplyr)
# Recode example
colon_s_small = colon_s %>%
  select(-id, -rx, -rx.factor) %>%
  mutate(
    age.factor2 = forcats::fct_collapse(age.factor,
      "<60 years" = c("<40 years", "40-59 years")),
    sex.factor2 = forcats::fct_recode(sex.factor,
    # Intentional miscode
      "F" = "Male",
      "M" = "Female")
  )
# Check
colon_s_small %>%
  finalfit::check_recode()

out = colon_s_small %>%
  select(-extent, -extent.factor,-time, -time.years) %>% # choose to exclude variables
  check_recode(include_numerics = TRUE)

## Recoding mydata$cinsiyet into mydata$Cinsiyet
mydata$Cinsiyet <- recode(mydata$cinsiyet,
               "K" = "Kadin",
               "E" = "Erkek")
mydata$Cinsiyet <- factor(mydata$Cinsiyet)

## Recoding mydata$tumor_yerlesimi into mydata$TumorYerlesimi
mydata$TumorYerlesimi <- recode(mydata$tumor_yerlesimi,
               "proksimal" = "Proksimal",
               "distal" = "Distal",
               "yaygın" = "Yaygin",
               "gö bileşke" = "GEJ",
               "antrum" = "Antrum")
mydata$TumorYerlesimi <- factor(mydata$TumorYerlesimi)

## Reordering mydata$TumorYerlesimi
mydata$TumorYerlesimi <- factor(mydata$TumorYerlesimi, levels=c("GEJ", "Proksimal", "Antrum", "Distal", "Yaygin"))

## Recoding mydata$histolojik_alt_tip into mydata$HistolojikAltTip
mydata$HistolojikAltTip <- recode(mydata$histolojik_alt_tip,
               "medüller benzeri" = "meduller benzeri")
mydata$HistolojikAltTip <- factor(mydata$HistolojikAltTip)

## Recoding mydata$lauren_siniflamasi into mydata$Lauren
mydata$Lauren <- recode(mydata$lauren_siniflamasi,
               "diffüz" = "diffuse",
               "???" = "medullary")
mydata$Lauren <- factor(mydata$Lauren)

## Recoding mydata$histolojik_derece into mydata$Grade
mydata$Grade <- recode(mydata$histolojik_derece,
               "az diferansiye" = "az",
               "iyi diferansiye" = "iyi",
               "orta diferansiye" = "orta")
mydata$Grade <- factor(mydata$Grade)

## Reordering mydata$Grade
mydata$Grade <- factor(mydata$Grade, levels=c("iyi", "orta", "az"))

mydata$Tstage <- stringr::str_match(mydata$patolojik_evre, paste('(.+)', "N", sep=''))[,2]

mydata$Nstage <- paste0("N",
    stringr::str_match(mydata$patolojik_evre, paste( "N", '(.+)', "M", sep=''))[,2]
    )

mydata$Mstage <- paste0("M", 
    stringr::str_match(mydata$patolojik_evre, paste("M", '(.+)', sep=''))[,2]
)

mydata <- mydata %>% 
    dplyr::mutate(
        T_stage = dplyr::case_when(
            grepl(pattern = "T1", x = .$Tstage) == TRUE ~ "T1",
            grepl(pattern = "T2", x = .$Tstage) == TRUE ~ "T2",
            grepl(pattern = "T3", x = .$Tstage) == TRUE ~ "T3",
            grepl(pattern = "T4", x = .$Tstage) == TRUE ~ "T4",
            TRUE ~ "Tx"
        )
    ) %>% 
dplyr::mutate(
        N_stage = dplyr::case_when(
            grepl(pattern = "N0", x = .$Nstage) == TRUE ~ "N0",
            grepl(pattern = "N1", x = .$Nstage) == TRUE ~ "N1",
            grepl(pattern = "N2", x = .$Nstage) == TRUE ~ "N2",
            grepl(pattern = "N3", x = .$Nstage) == TRUE ~ "N3",
            TRUE ~ "Nx"
        )
    ) %>% 
dplyr::mutate(
        M_stage = dplyr::case_when(
            grepl(pattern = "M0", x = .$Mstage) == TRUE ~ "M0",
            grepl(pattern = "M1", x = .$Mstage) == TRUE ~ "M1",
            TRUE ~ "Mx"
        )
    )

## Recoding mydata$cd44_oran into mydata$CD44
mydata$CD44 <- recode(mydata$cd44_oran,
               "2" = "positive",
               "0" = "negative",
               "1" = "negative",
               "3" = "positive")
mydata$CD44 <- factor(mydata$CD44)

## Recoding mydata$her2_skor into mydata$Her2
mydata$Her2 <- recode(mydata$her2_skor,
               "+3" = "3",
               "+1" = "1",
               "+2" = "2")
mydata$Her2 <- factor(mydata$Her2)
## Reordering mydata$Her2
mydata$Her2 <- factor(mydata$Her2, levels=c("0", "1", "2", "3"))

## Recoding mydata$msi into mydata$MMR
mydata$MMR <- recode(mydata$msi,
               "MSS" = "pMMR",
               "MSİ(PMS2,MLH1)" = "dMMR(PMS2,MLH1)",
               "MSİ(MSH2,MSH6)" = "dMMR(MSH2,MSH6)",
               "MSİ(PMS2)" = "dMMR(PMS2)")
mydata$MMR <- factor(mydata$MMR)

## Recoding mydata$msi into mydata$MMR2
mydata$MMR2 <- recode(mydata$msi,
               "MSS" = "pMMR",
               "MSİ(PMS2,MLH1)" = "dMMR",
               "MSİ(MSH2,MSH6)" = "dMMR",
               "MSİ(PMS2)" = "dMMR")
mydata$MMR2 <- factor(mydata$MMR2)

mydata <- mydata %>% 
    dplyr::mutate(
TumorPDL1gr1 = dplyr::case_when(
        t_pdl1 < 1 ~ "kucuk1",
        t_pdl1 >= 1 ~ "buyukesit1"
    )
    ) %>% 
dplyr::mutate(
TumorPDL1gr5 = dplyr::case_when(
        t_pdl1 < 5 ~ "kucuk5",
        t_pdl1 >= 5 ~ "buyukesit5"
    )
    )   %>% 
dplyr::mutate(
inflPDL1gr1 = dplyr::case_when(
        i_pdl1 < 1 ~ "kucuk1",
        i_pdl1 >= 1 ~ "buyukesit1"
    )
    ) %>% 
dplyr::mutate(
inflPDL1gr5 = dplyr::case_when(
        i_pdl1 < 5 ~ "kucuk5",
        i_pdl1 >= 5 ~ "buyukesit5"
    )
    )

## Recoding mydata$lvi into mydata$LVI
mydata$LVI <- recode(mydata$lvi,
               "var" = "Var",
               "yok" = "Yok")
mydata$LVI <- factor(mydata$LVI)
## Reordering mydata$LVI
mydata$LVI <- factor(mydata$LVI, levels=c("Yok", "Var"))

## Recoding mydata$pni into mydata$PNI
mydata$PNI <- recode(mydata$pni,
               "var" = "Var",
               "yok" = "Yok")
mydata$PNI <- factor(mydata$PNI)
## Reordering mydata$PNI
mydata$PNI <- factor(mydata$PNI, levels=c("Yok", "Var"))

## Recoding mydata$ln into mydata$LenfNoduMetastazi
mydata$LenfNoduMetastazi <- recode(mydata$ln,
               "var" = "Var",
               "yok" = "Yok")
mydata$LenfNoduMetastazi <- factor(mydata$LenfNoduMetastazi)
## Reordering mydata$LenfNoduMetastazi
mydata$LenfNoduMetastazi <- factor(mydata$LenfNoduMetastazi, levels=c("Yok", "Var"))

mydata$sontarih <- janitor::excel_numeric_to_date(as.numeric(mydata$olum_tarihi))

mydata$Outcome <- "Dead"
mydata$Outcome[mydata$olum_tarihi == "yok"] <- "Alive"

# cat(names(mydata), sep = ",\n")

mydata <- mydata %>% 
    select(
# sira_no,
# no,
# x3,
# hasta_biyopsi_no,
# cinsiyet,
        Cinsiyet,
        Yas = hasta_yasi,
        TumorYerlesimi,
        TumorCapi = tumor_capi,
HistolojikAltTip,
Lauren,
Grade,
TNM = patolojik_evre,
Tstage,
T_stage,
Nstage,
N_stage,
Mstage,
M_stage,
CD44,
Her2,
MMR,
MMR2,
TumorPDL1gr1,
TumorPDL1gr5,
inflPDL1gr1,
inflPDL1gr5,
LVI,
PNI,
LenfNoduMetastazi,
Outcome,        
# tumor_yerlesimi,

# histolojik_alt_tip,
# lauren_siniflamasi,
# histolojik_derece,
# cd44_oran,
# cd44_intense,
# her2_skor,
# msi,
# t_pdl1,
# i_pdl1,
# lvi,
# pni,
# ln,
CerrahiTarih = cerrahi_tarih,
# olum_tarihi,
genel_sagkalim,
SonTarih = sontarih
    )

iRenameColumn.R

iSelectColumn.R

mydata <- janitor::clean_names(mydata)
# cat(names(mydata), sep = ",\n")

names(mydata) <- c(names(mydata)[1:21], paste0("Soru", 1:30))

library(arsenal)
tab1 <- tableby(~ katilim_durumu
                ,
                data = mydata
)
summary(tab1)

mydata <- mydata %>% 
  filter(katilim_durumu == "katılmış ve tamamlamış")

# summarytools::view(summarytools::dfSummary(mydata))
# dplyr::glimpse(mydata)
# mydata %>%
#   select(starts_with("Soru")) %>% 
#   pivot_longer(everything()) %>% 
#   select(value) %>% 
#   pull() %>% 
#   unique() %>% 
#   cat(sep = "\n")

## Recoding mydata$x3_yasiniz_nedir into mydata$YasGrup
mydata$YasGrup <- factor(mydata$x3_yasiniz_nedir)
## Reordering mydata$YasGrup
mydata$YasGrup <- factor(mydata$YasGrup, levels=c("20-29", "30-39", "40-49", "50-59", "60-69", "70-79", "80-89"))

## Recoding mydata$x4_cinsiyetiniz_nedir into mydata$Cinsiyet
mydata$Cinsiyet <- recode(mydata$x4_cinsiyetiniz_nedir,
               "Kadın" = "Kadin")
mydata$Cinsiyet <- factor(mydata$Cinsiyet)

## Recoding mydata$x5_kac_yildir_genel_cerrahi_uzmanisiniz into mydata$UzmanlikSuresi
mydata$UzmanlikSuresi <- recode(mydata$x5_kac_yildir_genel_cerrahi_uzmanisiniz,
               "43739" = "10-19")
mydata$UzmanlikSuresi <- factor(mydata$UzmanlikSuresi)

## Reordering mydata$UzmanlikSuresi
mydata$UzmanlikSuresi <- factor(mydata$UzmanlikSuresi, levels=c("0-9", "10-19", "20-29", "30-39", "40-49"))

## Recoding mydata$x6_unvaniniz_nedir into mydata$Unvan
mydata$Unvan <- factor(mydata$x6_unvaniniz_nedir)

## Reordering mydata$Unvan
mydata$Unvan <- factor(mydata$Unvan, levels=c("Op.Dr.", "Doktor Öğretim Üyesi", "Doç.Dr.", "Prof.Dr"))

## Recoding mydata$x8_hangi_kurumda_calisiyorsunuz into mydata$Kurum
mydata$Kurum <- recode(mydata$x8_hangi_kurumda_calisiyorsunuz,
               "Eğitim Araştırma Hastanesi" = "Eğitim Araştırma",
               "İlçe Devlet Hastanesi" = "İlçe Devlet",
               "Üniversite Hastanesi" = "Üniversite",
               "İl Devlet Hastanesi" = "İl Devlet",
               "Özel Hastane ve Kurumlar" = "Özel")
mydata$Kurum <- factor(mydata$Kurum)

## Reordering mydata$Kurum
mydata$Kurum <- factor(mydata$Kurum, levels=c("Özel", "İlçe Devlet", "İl Devlet", "Eğitim Araştırma", "Üniversite"))

tersSorular <- c("Soru1",
                 "Soru4",
                 "Soru15",
                 "Soru17",
                 "Soru29")

CSS <- c(
  "Soru3",
  "Soru6",
  "Soru12",
  "Soru16",
  "Soru18",
  "Soru20",
  "Soru22",
  "Soru24",
  "Soru27",
  "Soru30"
)


BS <- c(
  "Soru1",
  "Soru4",
  "Soru8",
  "Soru10",
  "Soru15",
  "Soru17",
  "Soru19",
  "Soru21",
  "Soru26",
  "Soru29"
)


STSS <- c(
  "Soru2",
  "Soru5",
  "Soru7",
  "Soru9",
  "Soru11",
  "Soru13",
  "Soru14",
  "Soru23",
  "Soru25",
  "Soru28"
)

recode_numberize <- function(x, ...) {
  dplyr::recode(
    x,
    "Bazı zamanlar" = 3,
    "Çoksık" = 5,
    "Hiçbir zaman" = 1,
    "Nadiren" = 2,
    "Sık sık" = 4,
    "Sıkça" = 4,
"Bazı zamanlarda" = 3,
"Çok sık" = 5,
"Sıksık" = 4
    )
}


mydata <- mydata %>% 
    mutate_at(.tbl = .,
              .vars = vars(starts_with("Soru"), -tersSorular),
              .funs = recode_numberize
    )



recode_numberize_ters <- function(x, ...) {
  recode(
    x,
    "Bazı zamanlar" = 3,
"Çoksık" = 1,
"Hiçbir zaman" = 5,
"Nadiren" = 4,
"Sık sık" = 2,
"Sıkça" = 2,
"Bazı zamanlarda" = 3,
"Çok sık" = 1,
"Sıksık" = 2
        )
}


mydata <- mydata %>% 
    mutate_at(.tbl = .,
              .vars = vars(tersSorular),
              .funs = recode_numberize
    )

mydata <- mydata %>% 
  # böyle yazınca missing olunca hesaplamıyor
  # mutate(
  #   CSS_total = rowSums(select(., CSS), na.rm = FALSE)
  # ) %>% 
  mutate(
    CSS_total = rowSums(select(., CSS), na.rm = TRUE)
  ) %>% 
mutate(
    BS_total = rowSums(select(., BS), na.rm = TRUE)
  ) %>% 
  mutate(
    STSS_total = rowSums(select(., STSS), na.rm = TRUE)
  )


mydata <- mydata %>% 
  naniar::replace_with_na_at(
    .vars = vars(ends_with("_total")),
    condition = ~.x == 0
    )

<= 22 Low
>= 23 & <= 41 Average 
>=42 High

mydata <- mydata %>% 
  mutate_at(.tbl = .,
            .vars = vars(ends_with("_total")),
            .funs = list(Gr = 
                           ~ case_when(
                             . <= 22 ~ "Low",
                             . >= 23 & . <= 41 ~ "Average",
                             . >= 42 ~ "High",
                             TRUE ~ NA_character_
                           )
                           )

  ) %>% 
  mutate_at(.tbl = .,
            .vars = vars(ends_with("_Gr")),
            .funs = ~ factor(., levels=c("Low", "Average", "High"))
              )

# ## Reordering mydata$CSS_total_Gr
# mydata$CSS_total_Gr <- factor(mydata$CSS_total_Gr, )
# 
# ## Reordering mydata$BS_total_Gr
# mydata$BS_total_Gr <- factor(mydata$BS_total_Gr, levels=c("Low", "Average", "High"))
# 
# 
# ## Reordering mydata$STSS_total_Gr
# mydata$STSS_total_Gr <- factor(mydata$STSS_total_Gr, levels=c("Low", "Average", "High"))

sbalci/histopathology-template documentation built on June 29, 2023, 5:52 a.m.