生信入门马拉松之R语言基础-文件读取、导出和画图(day 5)

test <- read.csv("ex3.csv")#读取文件并赋值给test。
test
class("ex3.csv")#返回结果是字符串。
class(test)#判断test是什么数据类型。

#1.读取ex1.txt
ex1 <- read.table("ex1.txt")#行名和列名被归到第一行第一列，查找帮助文档有解决方案。
ex1[2,4]
ex1 <- read.table("ex1.txt",header = T)

#2.读取ex2.csv
ex2 <- read.csv("ex2.csv")#行名和列名没有正确归位，转化成矩阵时会发生数值型转变成字符型。
ex2
ex3 <- read.csv("ex2.csv",row.names = 1,check.names = F)#比较方便转化成矩阵

#注意：数据框不允许重复的行名

rod = read.csv("rod.csv",row.names = 1)#有重复值，无法设置第一行为行名
rod = read.csv("rod.csv")
?read.csv

#用data.table来读取,无法设置行名
library(data.table)
ex1 = fread("ex1.txt")
class(ex1)
ex1 = fread("ex1.txt",data.table = F)
class(ex1)
ex2 = fread("ex2.csv",data.table = F)
class(ex2)
#不支持直接设置行名
library(tibble)
ex2 = column_to_rownames(ex2,"V1")

#rio
library(rio)
#一个函数支持读取很多格式，见帮助文档
?rio
ex1 = import("ex1.txt")
class(ex1)
#一个函数支持导出很多格式，见帮助文档
export(ex1,file = "ex1.xlsx")

library(rio)
gz1 <- import("GSE217012_Normalized_RPKM_LOG2_matrix.txt.gz")
class(gz1)

load("y.Rdata")
class(y)#矩阵
y
mean(y$gene1)#出现报错,小tips：数据框如果有两种数据类型，转矩阵会出现问题。
y[,1]
mean(as.numeric(y[,1]))#正确，老师的讲解
y[,1] = as.numeric(y[,1])#运行后依然是字符型，因为矩阵只允许一种数据类型，没有办法修改某一列某一行甚至某个元素的数据类型，要改除非整个矩阵一起改。

#我的答案
export(y,file = "y.txt")
aa2 <- import("y.txt")
class(aa2)
mean(aa2$gene1)

#1.基础包 略显陈旧 了解一下
plot(iris[,1],iris[,3],col = iris[,5]) 
text(6.5,4, labels = 'hello')

dev.off() #关闭画板

#2.ggplot2 中坚力量，语法有个性
library(ggplot2)
ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,
                           y = Petal.Length,
                           color = Species))

#3.ggpubr 新手友好型 ggplot2简化和美化 褒贬不一
library(ggpubr)
ggscatter(iris,
          x="Sepal.Length",
          y="Petal.Length",
          color="Species")

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,y = Petal.Length),
             colors = "blue")

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,y = Petal.Length),
             size = 5,
             alpha = 0.5,
             shape = 8)
#size = 5：点的大小5；alpha = 0.5：透明度50%；shape = 8:点的形状。

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,y = Petal.Length),
             colors = "blue")#颜色设置设定，color是geom_point函数参数

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,y = Petal.Length，
                           colors=species)）
#颜色设置映射到iris数据框的species列，color是aes函数参数

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,
                           y = Petal.Length,
                           color = Species))+
  scale_color_manual(values = c("blue","grey","red"))

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,
                           y = Petal.Length,
                           color = Species))+
  scale_color_brewer(paletter = "set1")

### Q2-1 空心形状和实心形状都用color设置颜色
ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,
                           y = Petal.Length,
                           color = Species),
             shape = 17) #17号，实心的例子

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,
                           y = Petal.Length,
                           color = Species),
             shape = 2) #2号，空心的例子

### Q2-2 既有边框又有内心的，才需要color和fill两个参数

ggplot(data = iris)+
  geom_point(mapping = aes(x = Sepal.Length,
                           y = Petal.Length,
                           color = Species),
             shape = 24,
             fill = "black") #24号，双色的例子

load("test.Rdata")

ggplot(test)+
  geom_point(aes(x= a,y = b,color = change))

ggplot(test)+
  geom_point(aes(x= a,y = b,color = change))+
  scale_color_manual(values = c("darkgreen","grey","red"))