Question

我做了以下二项式glm：

fit_MLPAlog2 = glm(cbind(infected, not_infected)~treatment*log(day+1), family = binomial, data = data)

我想制作以下情节：

我试过了

plot(allEffects(mod = fit_MLPAlog2), type = "response", ylim = c(0,1),lwd = 0.5, ylab = "Above detection treshold", xlab = "Day", multiline = FALSE)

所以我得到以下图表：

我还试图用ggplot2绘制它，然后我得到数据点：

p1 <- ggplot(data, aes(day, propinfected, colour = treatment)) + 
      geom_point(size = 3) + geom_smooth(method="glm") + 
      labs(x = "Day", y = "Above treshold") + theme_bw() +
      theme(legend.position = "top", legend.text = element_text(size = 12), 
            legend.title = element_text(size = 14))

有人提示如何制作第一张情节？

以下是数据：

    > data
       treatment day infected not_infected propinfected
    1       CTRL   0        0           20         0.00
    2        DWV   0       20            0         1.00
    3       CTRL   4       11            9         0.55
    4        DWV   4       20            0         1.00
    5       CTRL   8       15            5         0.75
    6        DWV   8       18            2         0.90
    7       CTRL  12       16            4         0.80
    8        DWV  12       19            1         0.95
    9       CTRL  16       19            1         0.95
    10       DWV  16       19            1         0.95

提前致谢！

Answer 1

所以我觉得这很接近：

library(reshape2)

set.seed(1234)
rawdata <- data.frame(day = 1:16)
rawdata$CTRL <- c(0, 22,44,55, 56,62,68,  70,75,78,  82,84,87,  89,92,95)/100 + 
                                                               c(0,rnorm(15,0,0.03))
rawdata$DIW <- pmin(1.0,((160-rawdata$day)/170) + rnorm(16,0,0.02))
#rawdata$CTRLdots <- c(0,55,60,75,77)

data <- melt(rawdata,id.var=c("day"),value.name="propinfected",variable.name="treatment")

clrs <- c("CTRL"="grey","DIW"="grey")
ltyp <- c("CTRL"="solid","DIW"="dotted")
fils <- c("CTRL"="black","DIW"="white")

p1 <- ggplot(data, aes(day, propinfected, 
       colour = treatment,fill=treatment,linetype=treatment)) + 
  geom_point(size = 4,shape=21) + 
  geom_smooth(method="loess",fill=I("grey")) + 
  scale_y_continuous(limits=c(0,1)) +
  scale_color_manual(values=clrs)+
  scale_fill_manual(values=fils)+
  scale_linetype_manual(values=ltyp)+
  labs(x = "Day", y = "Above detection threshold (%)") + theme_bw() +
  theme(legend.text = element_text(size = 12), 
        legend.title = element_text(size = 14),
        legend.justification=c(1,0), 
        legend.position=c(1,0)
        )
  print(p1)

得到以下特性：

R：ggplot2在散点图和效果图之间叠加

1 个答案: