A code takes you to draw multi format sangjimei pictures such as interactive +pdf+png

Write it at the front

Sangjitu is very nice , It is used to observe the subordinate relationship between different categories , And draw a beautiful structure diagram , Of course, it can be used in many places . Here we use microbiome data phyloseq object , Quickly and easily build a sanggi map for you . So if you have phyloseq Try it directly . The test data in this paper comes from ggClusterNet, Everyone from github Download and use .

actual combat

Import requires R package

library(ggClusterNet)
library(tidyverse)
library(phyloseq)
library(tidyverse)
library(viridis)
library(patchwork)
library(networkD3)
data(ps)

We combined microbiological data at genus level , Then take the top 50 genera with the highest abundance , Then remove Unassigned”) %>%  vegan_tax() %>% Of microbes . Note that these functions exist in ggClusterNet in , Pay attention to download and installation .

tax = ps %>% 
  ggClusterNet::tax_glom_wt(ranks = 6) %>%
  filter_OTU_ps(50) %>%
  subset_taxa(Genus  != "Unassigned") %>%
  vegan_tax() %>%
  as.data.frame()
head(tax)

Build connections , A source —- One goal . Mark the labels of different classification levels .

id2 = c("k","p","c","o","f","g")
dat = NULL
for (i in 1:5) {
  dat <- tax[,c(i,i+1)] %>% distinct(.keep_all = TRUE) 
  colnames(dat) = c("source","target")
  dat$source = paste(id2[i],dat$source,sep = "_")
  dat$target = paste(id2[i+1],dat$target,sep = "_")
  if (i == 1) {
    dat2 = dat
  }  

  dat2 = rbind(dat2,dat)

}

dim(dat2)
# dat2 = dat2 %>% distinct(.keep_all = TRUE) 

head(dat2)

Structure abundance table , Here we are only right KO1 This single sample is enriched ：

otu = ps %>% 
  ggClusterNet::tax_glom_wt(ranks = 6) %>%
  scale_micro() %>%
  filter_OTU_ps(50) %>%
  subset_taxa(Genus  != "Unassigned") %>%
  vegan_otu() %>%
  t() %>%
  as.data.frame()

head(otu)
otutax = cbind(otu,tax)

id = rank.names(ps)[1:6]
dat = NULL
for (i in 1:6) {
  dat <- otutax %>%
    group_by(!!sym(id[i])) %>%
    summarise(sum(KO1))
  colnames(dat) = c("target","value")
  dat$target = paste(id2[i],dat$target,sep = "_")
  if (i == 1) {
    dat3 = dat
  }  

  dat3 = rbind(dat3,dat)
}

dat4 <- dat2 %>% left_join(dat3)
sankey = dat4

Construct nodes and connection tables , preservation html file ,pdf and png file .

nodes <- data.frame(name = unique(c(as.character(sankey$source),as.character(sankey$target))),stringsAsFactors = FALSE)
nodes$ID <- 0:(nrow(nodes)-1)
sankey <- merge(sankey,nodes,by.x = "source",by.y = "name")
sankey <- merge(sankey,nodes,by.x = "target",by.y = "name")
colnames(sankey) <- c("X","Y","value","source","target")
sankey <- subset(sankey,select = c("source","target","value"))
nodes <- subset(nodes,select = c("name"))

ColourScal='d3.scaleOrdinal() .range(["#E41A1C", "#377EB8", "#4DAF4A", "#984EA3", "#FF7F00", "#FFFF33", "#A65628", "#F781BF", "#999999"])'

sankey$energy_type <- sub(' .*', '', nodes[sankey$source + 1, 'name'])

p <- sankeyNetwork(Links = sankey, Nodes = nodes,
                   Source = "source",Target = "target",Value = "value",
                   NodeID = "name",
                   sinksRight=FALSE,
                   LinkGroup = 'energy_type',
                   colourScale= ColourScal, 
                   # nodeWidth=40,
                   # fontSize=13
                   # nodePadding=20
                   )
p

saveNetwork(p,"sankey1.html")
library(webshot)

webshot("sankey.html" , "sankey.png")
webshot("sankey.html" , "sankey.pdf")

Guess you like

iMeta brief introduction   Highly quoted articles   High color value drawing imageGP  Network analysis iNAP
iMeta Web tools   Metabolism group MetOrigin  Meggie cloud Lactation prediction DeepKla
iMeta review   Intestinal flora   Plant flora   Oral flora   Protein structure prediction

10000+： Flora analysis   Babies and dogs   Rhapsody of syphilis   carry DNA Hair Nature

Series of tutorials ： Introduction to microbiome  Biostar  Microbiome    Macro genome

Expertise ： Academic charts   High score article   Shengxin Scripture   An indispensable person

Article to read ： Macro genome   Parasite benefits   Evolutionary tree   Necessary skills ： put questions to   Search for   Endnote

Amplification analysis ： Chart interpretation   Analysis process   Statistical mapping

16S Function prediction    PICRUSt  FAPROTAX  Bugbase Tax4Fun

Biological science :   Intestinal bacteria   Life on the human body   Great leap forward in life    Cell warfare   The mystery of human body   

Written in the back

To encourage readers to communicate and quickly solve scientific research difficulties , We established “ Macro genome ” Discussion groups , There are already domestic and foreign 6000+ Researchers join . Please add editor in chief wechat meta-genomics Bring you into the group , Be sure to remark “ full name - Company - Research direction - The title / grade ”. Please indicate your identity for the senior title , There are also microorganisms at home and abroad PI Cooperation and exchange of group supply . Ask for help with technical problems , First read 《 How to ask questions gracefully 》 Learn how to solve problems , Unresolved intra group discussions , Don't talk about problems in private , Help colleagues .

Click to read the original text , Jump to the latest article catalog to read