跟着Nature学作图：R语言corrplot包画热图展示相关系数

用户7010445

发布于 2022-05-23 15:49:14

1.5K0

发布于 2022-05-23 15:49:14

文章被收录于专栏：小明的数据分析笔记本小明的数据分析笔记本

论文是

Environmental factors shaping the gut microbiome in a Dutch population

数据和代码的github主页链接

https://github.com/GRONINGEN-MICROBIOME-CENTRE/DMP

这个也是数据代码的下载链接，可以看目录结构

https://zenodo.org/record/5910709#.YmAcp4VBzic

今天的推文重复一下论文中的figure4b

image.png

首先是读取数据

plotCorMat <- read.table('health_disease_prediction/diseases_signature_correlation.txt',
                         header = T,
                         row.names = 1,
                         stringsAsFactors = F)
dim(plotCorMat)
plotCorMat[1:3,1:3]

对数据集的行名进行修改

rownames(plotCorMat) = c("1.Blood.Anemia",
                             "2.Blood.Thrombosis",
                             "3.Cancer.Any",
                             "4.Cardiovascular.Arrythmia.MedDiagnosed",
                             "5.Cardiovascular.Colesterol.high",
                             "6.Cardiovascular.Heart.Attack",
                             "7.Cardiovascular.Heart.Failure.Disorder",
                             "8.Cardiovascular.Heartrate.complains",
                             "9.Cardiovascular.Hypertension",
                             "10.Endocrine.DiabetesT2",
                             "11.Gastrointestinal.Stomach.Ulcer",
                             "12.Hepatologic.Gallstones",
                             "13.Mental.Any",
                             "14.Mental.Burn.Out",
                             "15.Mental.Depression",
                             "16.Mental.Other.anxiety",
                             "17.Mental.Panic.disorder",
                             "18.Neurological.Dizziness.Falling",
                             "19.Neurological.Mental.Fibromyalgia",
                             "20.Neurological.Migraine",
                             "21.Other.Autoimmune.Rheumatoid.Artritis",
                             "22.Other.Chronic.cystitis",
                             "23.Other.Chronic.Inflammation.Throatnose",
                             "24.Other.Chronic.Muscle.Weakness",
                             "25.Other.Fractures",
                             "26.Other.Incontinence",
                             "27.Other.Kidney.Stones",
                             "28.Other.Osteoarthritis",
                             "29.Other.Osteoporosis",
                             "30.Other.RSI",
                             "31.Pulmonary.Autoimmune.Asthma",
                             "32.Pulmonary.COPD",
                             "33.Skin.Autoimmune.Atopic.dermatitis",
                             "34.Skin.Autoimmune.Psoriasis",
                             "35.Skin.Autoimmune.Severe.acne",
                             "36.Gastrointestinal.Rome3_IBS.Any",
                             "37.None.NoDiseases"
)

对数据集的列名进行修改

colnames(plotCorMat) = 1:37

准备颜色

col1<-colorRampPalette(c( "#9B2226", "#AE2021","#BB3E03","#CA6702","#FFFFFF",
                          "#90e0ef", "#0077b6","#023e8a","#001219"))
col1(3)
col1(10)

作图代码

library(corrplot)
corrplot(as.matrix(plotCorMat),
         method = "square",
         tl.cex = 0.85,
         cl.cex = 0.85,
         col = col1(20),
         tl.col = "black")

这里 tl.cex是用来控制坐标轴文字的大小的cl.cex是用来控制图例刻度文字大小的

cl 是 colorlabel

image.png

这个是论文中提供的代码出图，和最终论文中用到的图还是有些差别的

下面我们查看corrplot这个包的帮助文档看看能够通过修改代码改成最终论文中的图的效果

如果需要把图例放到底部，直接添加一个cl.pos = "b"的参数

corrplot(as.matrix(plotCorMat),
         method = "square",
         tl.cex = 0.4,
         cl.cex = 0.4,
         col = col1(20),
         tl.col = "black",
         cl.pos = "b")