Question

我有一个闪亮的应用程序，可以在其中制作不同临床参数的ggplot。由于不同的临床参数包含不同的组，因此统计检验会有所不同。如何根据下拉菜单中的临床参数为stat_compare_means定义不同的组？

这是应用程序： https://mjellelab.shinyapps.io/SHINY_miRNA_CLINICAL/

例如，对于血清成分_m_component，我想将所有四个组彼此进行比较（未知除外），在ISS阶段，我要比较1vs2、1vs3和2vs3。

app：

data_prep <- 
structure(list(miRNA = c("hsa-let-7a-3p", "hsa-let-7a-3p", "hsa-let-7a-3p", 
"hsa-let-7a-3p", "hsa-let-7a-3p", "hsa-let-7a-3p"), ID = c("86", 
"175", "217", "394", "444", "618"), value = c(5.57979757386892, 
5.21619202802748, 5.42796072966512, -1.1390337316217, 5.06134249676025, 
4.37364284462968), Bone_disease = c("Without bone disease", "With bone disease", 
"With bone disease", "With bone disease", "With bone disease", 
"With bone disease"), Serum_M_component = structure(c(3L, 3L, 
5L, 3L, 1L, 3L), .Label = c("IgA", "IgD", "IgG", "LightChains", 
"Unknown"), class = "factor"), ISS_stage = c("Stage 3", "Stage 1", 
"Stage 3", "Stage 2", "Unknown", "Unknown")), row.names = c(NA, 
6L), class = "data.frame")


ui.miRNA.clinical <- dashboardPage(
  # Application title
  dashboardHeader(title=h4(HTML("MicroRNA expression <br/> in Multiple myeloma"))),
  dashboardSidebar(
    selectInput("p", "Clinical parameter", choices = c("Bone_disease", "Serum_M_component","ISS_stage")),
    selectInput("gene", "MicroRNA", choices = unique(data_prep$miRNA))),
  dashboardBody(
    tabsetPanel(
      tabPanel("Plot", plotOutput("myplot", width = "400px", height = "300px"))
    )
  )
)

server.miRNA.clinical <- function(input, output, session) {
  
  # filter data by Gene
  data_selected <- reactive({
    filter(data_prep, miRNA %in% input$gene)
  })
  

  output$myplot <- renderPlot({
    ggplot(data_selected(), aes_string(input$p, "value", fill = input$p)) + 
      geom_boxplot() + theme_classic(base_size = 12) + labs(x="Clinical parameter",y="MicroRNA expression (cpm,log2)") +
      stat_compare_means(method = "anova") 
      #method ="anova"
  })
}

stat_compare_means在闪亮的应用程序中-使用示例数据更新

0 个答案: