SAS绘图、R绘图哪个更适合你？ （二）：R绘图鉴赏
前面我们一起鉴赏了SAS的绘图本领，今天我们来领略R的绘图风采，两者各有千秋！后续将会一一更新绘图教程，包括图中的每个细节调整，关键的Code讲解，致力于共同学习进步，打造高质量图表！
下期我们将会就R和SAS两款统计分析软件的绘图功能进行对比，从绘图对于科研的重要性到绘图的原理，以及如何才算学会了绘图这门技巧进行详细讲解。
1.1 相关图
热图
pvalue <- matrix(ifelse(dfP <= 0.001, "***",
ifelse(dfP <= 0.01, "**",
ifelse(dfP <= 0.05, "*", " "))), nrow = nrow(dfspearman))
setwd("E:/DATA/TP/20230511-4500/FIG")
png("Heatmap4.png",units="in", width=10, height=8,res=400)
pheatmap(
dfspearman,
angle_col = 45,
color = colorRampPalette(c("#3C5488", "#FFFFFF", "#E64B35"))(50),
show_rownames = T,
display_numbers =pvalue,number_color="black",number_format="%.2e", fontsize_number = 16,
col_names_side = "top",
show_colnames =T,
fontfamily= "myFont",
border = F,
# cluster_rows = FALSE,
cluster_cols = FALSE,
fontsize = 16,
fontsize_row=16,
fontsize_col=16
)
dev.off()
网络图和弦图
library(showtext)
showtext_auto(enable = T)
nodegroup<-as.data.frame(cbind(Id=unique(c(df$x,df$y))))
#write.table(nodegroup,'node1.txt',row.names = F,sep = '\t',quote = F,fileEncoding = 'GB2312')
node1<-c('HRDB'= '#fb0d1c',
'VR'= '#b4701d','
2s'= '#c56cfb',
'pc'= '#c56cfb',
'pNd'= '#c56cfb',
'24e'= '#44fffc',
'rMf'= '#44fffc',
'rMSg'= '#44fffc',
'2s' ='#f8bf6e',
'Hk'= '#f8bf6e',
'Ho' ='#f8bf6e',
'24-LF'= '#fc665d',
'LF'= '#fc665d',
'L' ='#fc665d',
'24-LF/HF'= '#77d07a',
'LF/' ='#77d07a',
'24-SDNN' ='#77d07a',
'SDN' ='#78c7ec',
'SDNN' ='#78c7ec')
nodegroup$color1<-node1[nodegroup$Id]
pdf('circos2.pdf',width = 5,height = 6)
circos.clear()
circos.par("track.height" = 0.3, gap.after=c(5,5,rep(1,2),5,rep(1,2),5,rep(1,2),5,rep(1,2),5,rep(1,2),5,rep(1,1),5),
start.degree=90)
circos.initialize(factors =nodegroup$Id,xlim = c(0,1))
length(get.all.sector.index())
circos.track(ylim = c(0,1),bg.border=NA)
circos.trackPlotRegion(
track.index = 2,
ylim = c(0, 1),
track.height = 0.06, bg.border = 'black', bg.col =node1,
panel.fun = function(x, y) {
sector.index = get.cell.meta.data('sector.index')
xlim = get.cell.meta.data('xlim')
ylim = get.cell.meta.data('ylim')
circos.text(CELL_META$xcenter,
CELL_META$cell.ylim[2]+1 ,
CELL_META$sector.index,
facing = "reverse.clockwise",
adj = c(1,0),
niceFacing = TRUE,
cex = 0.6)
} )
nodecolor<-nodegroup$color1
names(nodecolor)<-nodegroup$Id
col_meth<- colorRamp2(c(0.1, 0.5, 0.9), c('#0b24fb','#feb7c2', '#fc0d1b'))
for(i in 1:nrow(df)){
circos.link(df$x[i], 0.5, df$y[i], 0.5,
lwd=df$r[i]*2,
col = col_meth(df$r[i]))
}
library(ComplexHeatmap)
lgd_links = Legend(at = c(0.1, 0.5, 0.9), col_fun = col_meth, title_position = "topleft",
title='Correlation')
lgd_list_vertical = packLegend(lgd_links)
draw(lgd_list_vertical, x = unit(4, "mm"), y = unit(4, "mm"), just = c("left", "bottom"))
dev.off()
基因互作图
1.2 聚类图
复杂热图
groupcolor <- c("#2878B5","#C82423")
names(groupcolor) <- c("pre","post") #color1
agecolor <- c("#008000","#3cb371","#84c09c","#5e8e7e")
names(agecolor) <- c("1","2","3","4") #color2
contentcolor <-c("#8ECFC9","#FFBE7A","#FA7F6F",'#82B0D2')
names(contentcolor) <- c("Th1","Th2","Th17/Th22","Multi") #color3
ann_colors <- list(Group=groupcolor, Content= contentcolor,Age=agecolor) #Color Summary List####heatmap####
library(pheatmap)windowsFonts(myFont = windowsFont("Times New Roman"), KT = windowsFont("楷体"),
ST = windowsFont("宋体"))
setwd("E:/DATA/TP/20230511-4500/FIG")
png("Heatmap3.png",units="in", width=10, height=8,res=400)
pheatmap( df, color = colorRampPalette(c("#3C5488", "#FFFFFF", "#E64B35"))(50),
show_rownames = T,
show_colnames =F,
annotation_row = dfgroup1,
annotation_col = dfgroup2,
fontfamily= "myFont",
annotation_colors = ann_colors,
border = F,
cluster_rows = FALSE,
cluster_cols = FALSE,
fontsize = 16,
fontsize_row=16,
fontsize_col=16
)
dev.off()
PCA可视化图
冷热点聚集性分析地图
1.3 展示模型中变量效应量的图
机器学习中特征度可视化
条件过程分析效应图
森林图
点线图
非线性模型
风险分数图
列线图
1.4 数据挖掘展示图
词云图
堆积图
箱线图
散点图
气泡图
1.5 机器学习
模型性能验证折线图
验证集区分度测试图
“黑盒”模型重要特征可视化
校正曲线与DCA曲线
setwd("D:/DATA/TP/20230527-COX-LASSO/FIG")
png("DCA2.png",units="in", width=6, height=6,res=400)
dca(Surv(Survival_time, Death) ~ Nomogram1,
data = df,
time =90,
thresholds = 1:100/ 100) %>%
plot(smooth = T,lwd = 1.2)+theme_classic()+
# theme_prism(base_size =17)+
theme(legend.position="top")+
scale_x_continuous(
limits = c(0, 0.85),
guide = "prism_minor")+
scale_colour_manual(values=c("blue","red","green","orange","grey"))+
theme(axis.text.x = element_text(size=25,face="bold",colour = "black",family="A"))+
theme(axis.text.y = element_text(size=25,face = "bold",colour = "black",family = "A"))+
theme(axis.title.x = element_text(size=25,face="bold",colour = "black",family="A"))+
theme(axis.title.y = element_text(size=25,face = "bold",colour = "black",family = "A"))
dev.off()
setwd("D:/DATA/TP/20230527-COX-LASSO/FIG")
png("DCA2.png",units="in", width=6, height=6,res=400)
dca(Surv(Survival_time, Death) ~ Nomogram0,
data = df,
time =30,
thresholds = 1:100/ 100) %>%
plot(smooth = T,lwd = 1.2)+theme_classic()+
# theme_prism(base_size =17)+
theme(legend.position="top")+
scale_x_continuous(
limits = c(0, 0.85),
guide = "prism_minor")+
scale_colour_manual(values= c("blue","red","green","orange","grey"))+ theme(axis.text.x = element_text(size=25,face="bold",colour = "black",family="A"))+
theme(axis.text.y = element_text(size=25,face = "bold",colour = "black",family = "A"))+
theme(axis.title.x = element_text(size=25,face="bold",colour = "black",family="A"))+
theme(axis.title.y = element_text(size=25,face = "bold",colour = "black",family = "A"))
dev.off()
常用科研
# 暖色调：#FF5733, #FFC300, #FFA012, #FF7F50, #FFB6C1
# 冷色调：#87CEFA, #00BFFF, #1E90FF, #6495ED, #ADD8E6
# 土色调：#8B4513, #A0522D, #D2691E, #CD853F, #F4A460
# 棕色调：#8B6B23, #BC8F8F, #DEB887, #D2B48C, #F5DEB3
# 紫色调：#9400D3, #8A2BE2, #9370DB, #BA55D3, #DA70D6
# 黄色调：#FFD700, #FFFF00, #F0E68C, #FFC107, #FFA500
# 绿色调：#00FF00, #32CD32, #3CB371, #2E8B57, #228B22
# 灰色调：#808080, #A9A9A9, #C0C0C0, #D3D3D3, #E6E6E6
关注微信公众号，获取更多相关内容！
以上绘图教程后续更新，如急需代码，请后台私信联系！
