SKKU BIOHRS-FINAL Review

Author

Kunhee Lee

R코드와 실행결과를 모두 아래 메일로 보내주세요.

lisalee1208@naver.com
jinseob2kim@gmail.com

필요한 library 및 데이터 불러오기

library(tableone);library(data.table);library(magrittr);library(survival);library(jstable)

url <- "https://raw.githubusercontent.com/jinseob2kim/lecture-snuhlab/master/data/example_g1e.csv"
dt <- fread(url, header=T)

# dt의 1~6행 살펴보기
head(dt)

   EXMD_BZ_YYYY RN_INDI HME_YYYYMM Q_PHX_DX_STK Q_PHX_DX_HTDZ Q_PHX_DX_HTN
1:         2009  562083     200909            0             0            1
2:         2009  334536     200911            0             0            0
3:         2009  911867     200903            0             0            0
4:         2009  183321     200908           NA            NA           NA
5:         2009  942671     200909           NA            NA           NA
6:         2009  979358     200912           NA            NA           NA
   Q_PHX_DX_DM Q_PHX_DX_DLD Q_PHX_DX_PTB Q_HBV_AG Q_SMK_YN Q_DRK_FRQ_V09N HGHT
1:           0            0           NA        3        1              0  144
2:           0            0           NA        2        1              0  162
3:           0            0           NA        3        1              0  163
4:          NA           NA           NA        3        1              0  152
5:          NA           NA           NA        3        1              0  159
6:          NA           NA           NA        2        1              0  157
   WGHT WSTC  BMI VA_LT VA_RT BP_SYS BP_DIA URN_PROT  HGB FBS TOT_CHOL  TG HDL
1:   61   90 29.4   0.7   0.8    120     80        1 12.6 117      264 128  60
2:   51   63 19.4   0.8   1.0    120     80        1 13.8  96      169  92  70
3:   65   82 24.5   0.7   0.6    130     80        1 15.0 118      216 132  55
4:   51   70 22.1   0.8   0.9    101     62        1 13.1  90      199 100  65
5:   50   73 19.8   0.7   0.8    132     78        1 13.0  92      162  58  40
6:   55   73 22.3   1.5   1.5    110     70        1 11.9 100      192 109  53
   LDL CRTN SGOT SGPT GGT GFR
1: 179  0.9   25   20  25  59
2:  80  0.9   18   15  28  74
3: 134  0.8   26   30  30  79
4: 114  0.9   18   14  11  61
5: 111  0.9   24   23  15  49
6: 117  0.7   15   12  14  83

Q1. Table 1 만들기(15점)

데이터(dt)의 연도 별 주어진 변수의 기술통계량을 Table로 나타내어라. 단, 변수 Q_SMK_YN과 Q_HBV_AG는 범주형으로.

주어진 변수 : “HGHT”, “WGHT”, “BMI”, “HDL”, “LDL”, “Q_SMK_YN”, “Q_HBV_AG” (총 7개)

# myVars : 주어진 변수 추출
myVars <- c("HGHT","WGHT","BMI","HDL","LDL", "Q_SMK_YN","Q_HBV_AG")

# catVars : 주어진 변수 중 범주형 변수(categorical variables) 추출
catVars <- c("Q_SMK_YN","Q_HBV_AG")

# 첫 번째 방법 : 범주형으로 미리 변환하지 않고 'factorVars'로 지정
CreateTableOne(vars= myVars, factorVars = catVars, strata= "EXMD_BZ_YYYY", data= dt)

                  Stratified by EXMD_BZ_YYYY
                   2009            2010           2011           2012          
  n                   214             236            223            234        
  HGHT (mean (SD)) 164.08 (9.13)   164.93 (8.73)  164.15 (9.58)  164.92 (9.20) 
  WGHT (mean (SD))  64.33 (12.50)   65.14 (12.24)  64.90 (12.84)  65.82 (12.41)
  BMI (mean (SD))   23.76 (3.39)    23.83 (3.27)   23.96 (3.46)   24.09 (3.34) 
  HDL (mean (SD))   55.61 (15.26)   55.16 (13.96)  57.31 (38.67)  55.95 (13.49)
  LDL (mean (SD))  150.95 (547.09) 112.99 (35.67) 112.94 (33.56) 117.53 (33.05)
  Q_SMK_YN (%)                                                                 
     1                125 (59.0)      132 (55.9)     140 (62.8)     146 (62.4) 
     2                 34 (16.0)       42 (17.8)      35 (15.7)      36 (15.4) 
     3                 53 (25.0)       62 (26.3)      48 (21.5)      52 (22.2) 
  Q_HBV_AG (%)                                                                 
     1                 14 ( 6.6)        6 ( 2.5)      12 ( 5.4)      12 ( 5.1) 
     2                129 (60.8)      147 (62.3)     147 (65.9)     160 (68.4) 
     3                 69 (32.5)       83 (35.2)      64 (28.7)      62 (26.5) 
                  Stratified by EXMD_BZ_YYYY
                   2013           2014           2015           p      test
  n                   243            254            240                    
  HGHT (mean (SD)) 164.91 (8.87)  164.32 (9.37)  164.48 (9.47)   0.890     
  WGHT (mean (SD))  64.91 (12.27)  64.47 (12.13)  66.08 (13.36)  0.705     
  BMI (mean (SD))   23.75 (3.33)   23.78 (3.32)   24.28 (3.54)   0.541     
  HDL (mean (SD))   55.46 (14.61)  55.64 (13.66)  56.27 (14.96)  0.935     
  LDL (mean (SD))  111.16 (32.33) 116.55 (65.55) 111.53 (31.83)  0.371     
  Q_SMK_YN (%)                                                   0.856     
     1                141 (58.0)     157 (61.8)     154 (64.2)             
     2                 35 (14.4)      38 (15.0)      36 (15.0)             
     3                 67 (27.6)      59 (23.2)      50 (20.8)             
  Q_HBV_AG (%)                                                   0.073     
     1                 10 ( 4.1)       9 ( 3.5)      14 ( 5.8)             
     2                168 (69.1)     189 (74.4)     162 (67.5)             
     3                 65 (26.7)      56 (22.0)      64 (26.7)

# 두 번째 방법 : 범주형으로 미리 변환 후 'vars'에 주어진 모든 변수 지정 
# catVars 범주형으로 변환
dt[, (catVars) := lapply(.SD, factor), .SDcols= catVars]

CreateTableOne(vars= myVars, strata= "EXMD_BZ_YYYY", data= dt)

                  Stratified by EXMD_BZ_YYYY
                   2009            2010           2011           2012          
  n                   214             236            223            234        
  HGHT (mean (SD)) 164.08 (9.13)   164.93 (8.73)  164.15 (9.58)  164.92 (9.20) 
  WGHT (mean (SD))  64.33 (12.50)   65.14 (12.24)  64.90 (12.84)  65.82 (12.41)
  BMI (mean (SD))   23.76 (3.39)    23.83 (3.27)   23.96 (3.46)   24.09 (3.34) 
  HDL (mean (SD))   55.61 (15.26)   55.16 (13.96)  57.31 (38.67)  55.95 (13.49)
  LDL (mean (SD))  150.95 (547.09) 112.99 (35.67) 112.94 (33.56) 117.53 (33.05)
  Q_SMK_YN (%)                                                                 
     1                125 (59.0)      132 (55.9)     140 (62.8)     146 (62.4) 
     2                 34 (16.0)       42 (17.8)      35 (15.7)      36 (15.4) 
     3                 53 (25.0)       62 (26.3)      48 (21.5)      52 (22.2) 
  Q_HBV_AG (%)                                                                 
     1                 14 ( 6.6)        6 ( 2.5)      12 ( 5.4)      12 ( 5.1) 
     2                129 (60.8)      147 (62.3)     147 (65.9)     160 (68.4) 
     3                 69 (32.5)       83 (35.2)      64 (28.7)      62 (26.5) 
                  Stratified by EXMD_BZ_YYYY
                   2013           2014           2015           p      test
  n                   243            254            240                    
  HGHT (mean (SD)) 164.91 (8.87)  164.32 (9.37)  164.48 (9.47)   0.890     
  WGHT (mean (SD))  64.91 (12.27)  64.47 (12.13)  66.08 (13.36)  0.705     
  BMI (mean (SD))   23.75 (3.33)   23.78 (3.32)   24.28 (3.54)   0.541     
  HDL (mean (SD))   55.46 (14.61)  55.64 (13.66)  56.27 (14.96)  0.935     
  LDL (mean (SD))  111.16 (32.33) 116.55 (65.55) 111.53 (31.83)  0.371     
  Q_SMK_YN (%)                                                   0.856     
     1                141 (58.0)     157 (61.8)     154 (64.2)             
     2                 35 (14.4)      38 (15.0)      36 (15.0)             
     3                 67 (27.6)      59 (23.2)      50 (20.8)             
  Q_HBV_AG (%)                                                   0.073     
     1                 10 ( 4.1)       9 ( 3.5)      14 ( 5.8)             
     2                168 (69.1)     189 (74.4)     162 (67.5)             
     3                 65 (26.7)      56 (22.0)      64 (26.7)

Q2. 선형회귀, 로지스틱, 콕스생존분석 Table 만들기(15점)

2-1. 선형 회귀분석(Linear regression)(5점)

survival 패키지에 내장되어 있는 대장암 데이터 ‘colon’ 사용
time ~ rx + age + sex 선형회귀 실행 후 table로 나타내어라.

# colon의 1~6행 살펴보기
head(colon)

  id study      rx sex age obstruct perfor adhere nodes status differ extent
1  1     1 Lev+5FU   1  43        0      0      0     5      1      2      3
2  1     1 Lev+5FU   1  43        0      0      0     5      1      2      3
3  2     1 Lev+5FU   1  63        0      0      0     1      0      2      3
4  2     1 Lev+5FU   1  63        0      0      0     1      0      2      3
5  3     1     Obs   0  71        0      0      1     7      1      2      2
6  3     1     Obs   0  71        0      0      1     7      1      2      2
  surg node4 time etype
1    0     1 1521     2
2    0     1  968     1
3    0     0 3087     2
4    0     0 3087     1
5    0     1  963     2
6    0     1  542     1

res.reg <- glm(time ~ rx + age + sex, data = colon)
tb.reg <- glmshow.display(res.reg)     # 'jstable 패키지의 glmshow.display' 이용
knitr::kable(tb.reg$table, caption = tb.reg$first.line)

Linear regression predicting time
	crude coeff.(95%CI)	crude P value	adj. coeff.(95%CI)	adj. P value
rx: ref.=Obs	NA	NA	NA	NA
Lev	24.66 (-79.49,128.82)	0.643	22.98 (-81.3,127.27)	0.666
Lev+5FU	271.07 (166.41,375.74)	< 0.001	273.05 (168.19,377.9)	< 0.001
age	0.38 (-3.22,3.99)	0.835	0.37 (-3.21,3.95)	0.84
sex	13.93 (-72.26,100.12)	0.751	32.67 (-53.21,118.55)	0.456

2-2. 로지스틱 회귀분석(Logistic regression)(5점)

마찬가지로 대장암 데이터 ‘colon’ 사용
status ~ rx + age + sex 로지스틱 회귀 실행 후 table로 나타내어라.

res.logistic <- glm(status ~ rx + age + sex, data = colon, family = binomial)
tb.logistic <- glmshow.display(res.logistic)   # 'jstable 패키지의 glmshow.display' 이용
knitr::kable(tb.logistic$table, caption = tb.logistic$first.line)

Logistic regression predicting status
	crude OR.(95%CI)	crude P value	adj. OR.(95%CI)	adj. P value
rx: ref.=Obs	NA	NA	NA	NA
Lev	0.96 (0.77,1.2)	0.709	0.96 (0.77,1.2)	0.747
Lev+5FU	0.55 (0.44,0.68)	< 0.001	0.54 (0.43,0.68)	< 0.001
age	1 (0.99,1)	0.296	1 (0.99,1)	0.294
sex	0.97 (0.81,1.17)	0.758	0.93 (0.77,1.12)	0.454

2-3. 콕스 생존분석(Cox proportional hazard)(5점)

Surv(time, status) ~ rx + age + sex 실행 후 table로 나태내어라.

res.cox <- coxph(Surv(time, status) ~ rx + age + sex, data = colon, model = T)
tb.cox <- cox2.display(res.cox)   # 'jstable 패키지의 cox2.display' 이용
knitr::kable(tb.cox$table, caption = tb.cox$caption)

Cox model on time (‘time’) to event (‘status’)
	crude HR(95%CI)	crude P value	adj. HR(95%CI)	adj. P value
rx: ref.=Obs	NA	NA	NA	NA
Lev	0.98 (0.84,1.14)	0.786	0.98 (0.84,1.14)	0.811
Lev+5FU	0.64 (0.55,0.76)	< 0.001	0.64 (0.55,0.76)	< 0.001
age	1 (0.99,1)	0.382	1 (0.99,1)	0.468
sex	0.97 (0.85,1.1)	0.61	0.95 (0.84,1.08)	0.446

=> 2-1, 2-2, 2-3 table 3개를 Rmarkdown으로 제출하시오.

Q3. 샤이니 웹 앱 만든 후 ShinyApps.io 배포하기(10점)

지난 주 강의 때 실습한 NBA 2018/19 시즌 스탯 혹은 직접 제작한 샤이니 웹 앱 제작하기
ShinyApps.io를 통해 배포한 후, 생성된 url도 함께 제출하시오.

# 참고용 예시 코드

library(shiny)
library(ggplot2)
library(dplyr)
library(DT)

players <- read.csv("data/nba2018.csv")

ui <- fluidPage(
  titlePanel("NBA 2018/19 Player Stats"),
  sidebarLayout(
    sidebarPanel(
      "Exploring all player stats from the NBA 2018/19 season",
      h3("Filters"),
      sliderInput(
        inputId = "VORP",
        label = "Player VORP rating at least",
        min = -3, max = 10,
        value = 0
      ),
      selectInput(
        "Team", "Team",
        unique(players$Team),
        selected = "Golden State Warriors"
      )
    ),
    mainPanel(
      strong(
        "There are",
        textOutput("num_players", inline = TRUE),
        "players in the dataset"
      ),
      plotOutput("nba_plot"),
      DTOutput("players_data")
    )
  )
)

server <- function(input, output, session) {

  output$players_data <- renderDT({
    data <- players %>%
      filter(VORP >= input$VORP,
             Team %in% input$Team)

    data
  })

  output$num_players <- renderText({
    data <- players %>%
      filter(VORP >= input$VORP,
             Team %in% input$Team)

    nrow(data)
  })

  output$nba_plot <- renderPlot({
    data <- players %>%
      filter(VORP >= input$VORP,
             Team %in% input$Team)

    ggplot(data, aes(Salary)) +
      geom_histogram()
  })

}

shinyApp(ui, server)

반드시 모든 문제의 ’R 코드’와 ’실행결과(Table, Rmarkdown, url)’를 보내주세요.
시험/수업 관련 질문은 조교 메일(lisalee1208@naver.com)로 보내 주시면 감사하겠습니다.
한 학기 동안 수고 많으셨습니다 :)