@TOC

一、群体药代动力学数据探索目的和方法

数据探索目的

建模前通过图法和统计学方法，对研究数据探索可揭示数据特征，发现明显趋势，辨识离群值，异常值等。

数据来源

数据来自于焦正主编的基础群体药动学和药效学书籍第10章第二节模型数据

方法

作图主要用ggplot2包，及代码中涉及的包。

作图所用主题为ggplot2默认主题：theme_bw() 稍调整

二、数据探索过程

PK数据探索

加载的包和设目录

代码：

#加载的包----
library(ggplot2)
library(dplyr)
library(scales)

#设置目录----
x<-readline() #读取以下目录字段
D:\study\NONMEN\JIAOZHENG\第10章第二节代码示例199\第10章第二节代码示例\chapter10-section2\datacheck  
x <- gsub("\\\\", "/", x) # 将反斜杠转换为斜
setwd(x)  #设置目录

读取数据和数据处理

代码：

#数据读取和处理----
DATA<-read.csv("10-2.csv",header=T,skip=0) #读取数据：数据为焦正主编的基础群体药动学和药效学书籍第10章第二节模型数据
DATA$GROUP<-paste(DATA$DOSE,"mg")   #组别加上单位
DATA$DOSE_f<-factor(DATA$GROUP,levels=c("2 mg","3 mg","4 mg","5 mg"))   #组别转化分类因子

PKDATA<-filter(DATA,MDV==0,DVID==1)  #提取PK数据
PDDATA<-filter(DATA,MDV==0,DVID==2)  #提取PD数据

PK数据结果(部分)：
温故知新：基于R语言的群体药代动力学数据探索
PD数据结果(部分)：

平均血药浓度计算和作图

计算血药浓度均值代码：

#计算血药浓度均值和标准差----
ConcvsTIMEALL<-summarise(group_by(PKDATA,TIME,DOSE_f), #按照分类计算
                         mean=mean(DV),   #计算均值
                         sd=sd(DV),       #计算标准差
                         n=sum(!is.na(DV)),
                         se=sd/sqrt(n))   #计算标准误差

计算结果：
温故知新：基于R语言的群体药代动力学数据探索
平均血药浓度图（每个剂量组情况）

代码：

#作平均血药浓度图(按照组别分组）----
CONCPprofile<-
  ggplot()+
  geom_point(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f)))+   #数据-点
  geom_line(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f),group=DOSE_f),size=0.7)+ #数据-线
  labs(color="",shape="")+
  xlab("Time(day)")+ylab("Concentration (ng/mL)")+   #坐标轴标题
  # xlim(0,216)+  #x轴限定
  # ylim(0,20)+   #y轴限定
  scale_x_continuous(breaks=c(c(0,48,96,144,192,240,288,336,384,432,480,528)),
                     labels=c("0","2","4","6","8","10","12","14","16","18","20","22"))+ #坐标轴刻度设置
  facet_wrap(~DOSE_f,ncol=2)+   #分两列
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='red'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCPprofile
ggsave("ALLbygroupprofile.jpeg", CONCPprofile, width = 6, height = 4)

CONCPprofile<-
  ggplot()+
  geom_point(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f)))+
  geom_line(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f),group=DOSE_f),size=0.7)+
  labs(color="",shape="")+
  xlab("Time(day)")+ylab("Concentration (ng/mL)")+
  # xlim(0,216)+
  # ylim(0,20)+
  # scale_shape_manual(values = c(4:10))+
  scale_x_continuous(breaks=c(c(0,48,96,144,192,240,288,336,384,432,480,528)),
                     labels=c("0","2","4","6","8","10","12","14","16","18","20","22"))+
  facet_wrap(~DOSE_f,ncol=2)+
  scale_y_log10()+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='red'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCPprofile
ggsave("ALLbygroupprofilelog.jpeg", CONCPprofile, width = 6, height = 4)

作图结果

正常坐标：
温故知新：基于R语言的群体药代动力学数据探索
对数坐标：

作平均血药浓度图(总体情况）

代码：

#作平均血药浓度图(总体）----
CONCPprofile<-
  ggplot()+
  geom_point(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f)))+
  geom_line(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f),group=DOSE_f),size=0.7)+
  labs(color="",shape="")+
  xlab("Time(day)")+ylab("Concentration (ng/mL)")+
  # xlim(0,216)+
  # ylim(0,20)+
  # scale_shape_manual(values = c(4:10))+
  scale_x_continuous(breaks=c(c(0,48,96,144,192,240,288,336,384,432,480,528)),
                     labels=c("0","2","4","6","8","10","12","14","16","18","20","22"))+

  # facet_wrap(~DOSE_f,ncol=2)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='red'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCPprofile
ggsave("ALLgroupprofile.jpeg", CONCPprofile, width = 6, height = 4)

CONCPprofile<-
  ggplot()+
  geom_point(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f)))+
  geom_line(data=ConcvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f),group=DOSE_f),size=0.7)+
  labs(color="",shape="")+
  xlab("Time(day)")+ylab("Concentration (ng/mL)")+
  # xlim(0,216)+
  # ylim(0,20)+
  scale_x_continuous(breaks=c(c(0,48,96,144,192,240,288,336,384,432,480,528)),
                     labels=c("0","2","4","6","8","10","12","14","16","18","20","22"))+
  scale_y_log10()+
  theme_test()+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='red'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCPprofile
ggsave("ALLgroupprofilelog.jpeg", CONCPprofile, width = 6, height = 4)

作图结果

正常坐标：
温故知新：基于R语言的群体药代动力学数据探索
对数坐标：

每个受试者血药浓度图

SAD受试者血药浓度

代码

#作SADCONC受试者分组图（1-72受试者）----
CONCIDprofile<-
  ggplot(data=filter(PKDATA,CID<73,TIME<=24),aes(x=TIME))+
  geom_line(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=1)+
  geom_point(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # geom_line(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=2)+
  # geom_point(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # xlim(0,24)+
  facet_wrap(~CID,ncol=8)+
  xlab("\nTime after Dose (h)")+ylab("Concentration (ng/mL)\n")+
  guides(color=FALSE)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='black'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCIDprofile
ggsave("SADCONCIDprofile1.jpeg", CONCIDprofile, width = 8, height = 9)

#作SADCONC受试者分组图（72-144受试者）----
CONCIDprofile<-
  ggplot(data=filter(PKDATA,CID>72,TIME<=24),aes(x=TIME))+
  geom_line(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=1)+
  geom_point(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # geom_line(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=2)+
  # geom_point(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # xlim(0,24)+
  # scale_y_log10()+
  facet_wrap(~CID,ncol=8)+
  xlab("\nTime after Dose (h)")+ylab("Concentration (ng/mL)\n")+
  guides(color=FALSE)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='black'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCIDprofile
ggsave("SADCONCIDprofile2.jpeg", CONCIDprofile, width = 8, height = 9)

作图结果
温故知新：基于R语言的群体药代动力学数据探索

MAD受试者血药浓度

代码

#作MADCONC受试者分组图（1-72受试者）--------
CONCIDprofile<-
  ggplot(data=filter(PKDATA,CID<73,TIME>=480),aes(x=TIME))+
  geom_line(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=1)+
  geom_point(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # geom_line(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=2)+
  # geom_point(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # xlim(0,24)+
  facet_wrap(~CID,ncol=8)+
  xlab("\nTime after Dose (h)")+ylab("Concentration (ng/mL)\n")+
  guides(color=FALSE)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='black'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCIDprofile
ggsave("MADCONCIDprofile1.jpeg", CONCIDprofile, width = 8, height = 9)

#作MADCONC受试者分组图（72-144受试者）----
CONCIDprofile<-
  ggplot(data=filter(PKDATA,CID>72,TIME>=480),aes(x=TIME))+
  geom_line(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=1)+
  geom_point(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # geom_line(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=2)+
  # geom_point(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # xlim(0,24)+
  # scale_y_log10()+
  facet_wrap(~CID,ncol=8)+
  xlab("\nTime after Dose (h)")+ylab("Concentration (ng/mL)\n")+
  guides(color=FALSE)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='black'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
CONCIDprofile
ggsave("MADCONCIDprofile2.jpeg", CONCIDprofile, width = 8, height = 9)

作图结果
温故知新：基于R语言的群体药代动力学数据探索

PD数据探索

PD数据探索类似于PK数据探索，故只探索部分

平均药效指标图

代码

#作平均药效指标图(按照组别分组）----
PDPprofile<-
  ggplot()+
  geom_point(data=PDvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f)))+
  geom_line(data=PDvsTIMEALL,aes(x=TIME,y=mean,color=factor(DOSE_f),group=DOSE_f),size=0.7)+
  labs(color="",shape="")+
  xlab("Time(day)")+ylab("Concentration (ng/mL)")+
  # xlim(0,216)+
  # ylim(0,20)+
  # scale_shape_manual(values = c(4:10))+
  scale_x_continuous(breaks=c(c(0,48,96,144,192,240,288,336,384,432,480,528)),
                     labels=c("0","2","4","6","8","10","12","14","16","18","20","22"))+

  facet_wrap(~DOSE_f,ncol=2)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='black'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
PDPprofile
ggsave("ALLPDbygroupprofile.jpeg", PDPprofile, width = 6, height = 4)

作图结果
温故知新：基于R语言的群体药代动力学数据探索

每个受试者药效指标图

代码

#作每个ID的药效指标图----
PDIDprofile<-
  ggplot(data=filter(PKDATA,CID<73,TIME>=480),aes(x=TIME))+
  geom_line(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=1)+
  geom_point(aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # geom_line(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),linetype=2)+
  # geom_point(data=filter(PKDATASAD,fast==0),aes(y=DV,group=DOSE_f,color=factor(DOSE_f)),shape=21,size=1)+
  # xlim(0,24)+
  facet_wrap(~CID,ncol=8)+
  xlab("\nTime after Dose (h)")+ylab("Concentration (ng/mL)\n")+
  guides(color=FALSE)+
  theme_bw()+  #内置主题
  theme(axis.text=element_text(size=10,color='black'), #坐标轴字体设置
        axis.title = element_text(color='black',face = "bold"),#坐标轴标题设置
        panel.grid.major =element_blank(), panel.grid.minor = element_blank(),#网格线设置
        legend.position = "right") #图例设置
PDIDprofile
ggsave("SADPDIDprofile.jpeg", PDIDprofile, width = 8, height = 6)

作图结果
温故知新：基于R语言的群体药代动力学数据探索

评论（23 条评论）

free binance account

2025-12-21 08:19:59 回复

Your article helped me a lot, is there any more related content? Thanks! https://www.binance.info/register?ref=IXBIAFVY

Windows Edge 中国广西玉林市电信

create a binance account

2026-01-10 20:18:03 回复

Thank you for your sharing. I am worried that I lack creative ideas. It is your article that makes me full of hope. Thank you. But, I have a question, can you help me?

Josephbic

2026-01-14 03:34:31 回复

我经常阅读出行博客。太棒了看到照片。奧運聖火能感受到热爱。由衷感谢温暖。

Windows Chrome 乌克兰

Registrasi

2026-01-15 20:01:37 回复

Thanks for sharing. I read many of your blog posts, cool, your blog is very good.

Binance

2026-01-18 03:17:07 回复

Code Binance

2026-01-20 15:35:40 回复

binance

2026-01-31 15:31:32 回复

Your point of view caught my eye and was very interesting. Thanks. I have a question for you. https://www.binance.com/es-MX/register?ref=GJY4VW8W

binance register

2026-02-06 12:03:03 回复

Your point of view caught my eye and was very interesting. Thanks. I have a question for you. https://www.binance.com/ur/register?ref=SZSSS70P

binance Konto er"offnen

2026-02-09 06:05:42 回复

Your enticle helped me a lot, is there any more related content? Thanks! https://accounts.binance.info/lv/register?ref=SMUBFN5I

Inscription Binance

2026-02-12 13:37:47 回复

Can you be more specific about the content of your article? After reading it, I still have some doubts. Hope you can help me.

Binance Pagpaparehistro

2026-02-16 16:11:24 回复

Your article helped me a lot, is there any more related content? Thanks!

2026-02-25 20:41:54 回复

Buat Akun Binance

2026-03-05 19:58:54 回复

Thank you for your sharing. I am worried that I lack creative ideas. It is your article that makes me full of hope. Thank you. But, I have a question, can you help me? https://www.binance.info/register?ref=IXBIAFVY

binance Anmeldebonus

2026-03-06 19:29:05 回复

Your point of view caught my eye and was very interesting. Thanks. I have a question for you.

Registrera

2026-03-10 20:35:22 回复

2026-03-20 08:33:46 回复

Can you be more specific about the content of your article? After reading it, I still have some doubts. Hope you can help me. https://accounts.binance.com/register-person?ref=JW3W4Y3A

Windows Edge 中国广西钦州市电信

Inscription à Binance US

2026-04-02 00:12:16 回复

Thanks for sharing. I read many of your blog posts, cool, your blog is very good. https://accounts.binance.info/zh-CN/register-person?ref=WFZUU6SI

Binance注册

2026-04-06 08:17:41 回复

binance sign up

2026-04-11 03:38:49 回复

Can you be more specific about the content of your article? After reading it, I still have some doubts. Hope you can help me. https://accounts.binance.com/fr/register?ref=T7KCZASX

2026-04-15 17:24:55 回复

Your point of view caught my eye and was very interesting. Thanks. I have a question for you. https://accounts.binance.com/es-AR/register?ref=UT2YTZSU

Create a free account

2026-04-15 18:29:24 回复

Binance创建账户

2026-04-30 22:50:14 回复

binance h"anvisning

2026-05-15 02:10:55 回复

I don’t think the title of your article matches the content lol. Just kidding, mainly because I had some doubts after reading the article.

Windows Edge 中国广西百色市电信

温故知新：基于R语言的群体药代动力学数据探索

一 、群体药代动力学数据探索目的和方法

数据探索目的

数据来源

方法

二、数据探索过程

PK数据探索

加载的包和设目录

读取数据和数据处理

平均血药浓度计算和作图

每个受试者血药浓度图

PD数据探索

平均药效指标图

每个受试者药效指标图

一、群体药代动力学数据探索目的和方法