条形图分类数据综述

Question

I ^有一个如下所示的数据集：

    typestudy   dloop cytb coi  other microsat  SNP
    methods     no  no  no  no  yes no
    methods     yes no  no  no  no  yes
    methods     no  no  no  no  yes no
    methods     no  no  no  no  yes no
    wildcrime   no  no  no  yes no  no
    taxonomy    no  no  no  no  yes no
    methods     yes no  no  no  no  no
    methods     no  no  no  no  yes no
    taxonomy    no  no  no  no  yes no
    wildcrime   yes no  no  no  no  no
    methods     yes no  no  no  no  no
    taxonomy    no  no  no  no  yes yes
    taxonomy    no  no  no  no  yes no

但它有10列是/否，与进一步的遗传元素相对应，有200多行。

在Excel中，图形摘要选项给了我一个很好的堆叠条形图，但我需要能够在R中重新创建它，以满足我的报告的大学标准

    > summary(dframe1$type.of.study)
         methods development                        other 
                          49                            5 
population genetic structure                     taxonomy 
                          91                           86 
              wildlife crime 
                           6 
    > barplot(as.matrix(dframe1))
     There were 11 warnings (use warnings() to see them)
    > warnings()
    Warning messages:
    1: In apply(height, 2L, cumsum) : NAs introduced by coercion
    2: In apply(height, 2L, cumsum) : NAs introduced by coercion
    3: In apply(height, 2L, cumsum) : NAs introduced by coercion
    4: In apply(height, 2L, cumsum) : NAs introduced by coercion
    5: In apply(height, 2L, cumsum) : NAs introduced by coercion
    6: In apply(height, 2L, cumsum) : NAs introduced by coercion
    7: In apply(height, 2L, cumsum) : NAs introduced by coercion
    8: In apply(height, 2L, cumsum) : NAs introduced by coercion
    9: In apply(height, 2L, cumsum) : NAs introduced by coercion
   10: In apply(height, 2L, cumsum) : NAs introduced by coercion
   11: In apply(height, 2L, cumsum) : NAs introduced by coercion

这给了我这个

​

​

我也成功地制作了这个，但是找不到我所用的脚本

​

​

我的目标与此类似：

​

​

这是相当可悲的，但我花了将近一个星期的故障排除基于在线资源和其他问题在这里达到这一点。我不知道如何计算研究的类型，所以他们被计算出来，使不同的遗传标记对应的条形图的高度。我知道这对于堆栈溢出的标准来说太模糊了，但是我很绝望，所以我不想再提这个了，以防有人有任何建议

^ (我会尽量简洁，但你会看到我在R方面的流利程度太差了，我不会寻求帮助，但我花了好几天时间处理这些数据，我被吓呆了，我永远找不到解决办法，也没有人在我的研究职位上寻求帮助)

Answer 1

我们可以通过循环获得每一列的table，然后执行barplot。

barplot(sapply(df1[-1], function(x) table(factor(x, 
     levels = c("yes", "no")))), col = c("red", "blue"))
legend("topright", legend = c("yes", "no"), fill = c("red", "blue"))

Answer 2

以防万一你想要的东西看起来像这个例子(有点)：

df <- read.table(text = "typestudy   dloop cytb coi  other microsat  SNP
    methods     no  no  no  no  yes no
                 methods     yes no  no  no  no  yes
                 methods     no  no  no  no  yes no
                 methods     no  no  no  no  yes no
                 wildcrime   no  no  no  yes no  no
                 taxonomy    no  no  no  no  yes no
                 methods     yes no  no  no  no  no
                 methods     no  no  no  no  yes no
                 taxonomy    no  no  no  no  yes no
                 wildcrime   yes no  no  no  no  no
                 methods     yes no  no  no  no  no
                 taxonomy    no  no  no  no  yes yes
                 taxonomy    no  no  no  no  yes no", 
                 header = T, stringsAsFactors = F)

library(tidyr)
library(ggplot2)
library(dplyr)
df %>% gather(key = key, value = value, -typestudy) %>% 
  filter(value == "yes") %>% 
  ggplot(aes(x = key, fill = typestudy)) +
  geom_bar() + 
  coord_flip() + 
  theme_minimal() +
  theme(legend.position = "bottom",
        panel.grid.minor = element_blank(),
        panel.grid.major.y = element_blank()) +
  xlab(NULL) +
  ylab(NULL)

Answer 3

我不知道您是否只是在yeses之后，但是这里有一种可能，可以让您使用no，以防您想要按响应类型(是/否)的巴图。

df %>%
  gather(var, value, -typestudy) %>%
  group_by(typestudy, var, value) %>%
  count() %>%
  filter(value == "yes") %>%
  ggplot(aes(var, n, group = typestudy, fill = typestudy)) +
  geom_bar(stat = "identity") +
  scale_fill_brewer(palette = "Dark2", direction = -1) +
  coord_flip() +
  theme(
    axis.title.x=element_blank(),
    axis.title.y=element_blank(),
    legend.position = "bottom",
    panel.grid.minor = element_blank(),
    panel.grid.major.y = element_blank(),
    legend.title=element_blank())

​

​

数据

df <- structure(list(typestudy = c("methods", "methods", "methods", 
"methods", "wildcrime", "taxonomy", "methods", "methods", "taxonomy", 
"wildcrime", "methods", "taxonomy", "taxonomy"), dloop = c("no", 
"yes", "no", "no", "no", "no", "yes", "no", "no", "yes", "yes", 
"no", "no"), cytb = c("no", "no", "no", "no", "no", "no", "no", 
"no", "no", "no", "no", "no", "no"), coi = c("no", "no", "no", 
"no", "no", "no", "no", "no", "no", "no", "no", "no", "no"), 
    other = c("no", "no", "no", "no", "yes", "no", "no", "no", 
    "no", "no", "no", "no", "no"), microsat = c("yes", "no", 
    "yes", "yes", "no", "yes", "no", "yes", "yes", "no", "no", 
    "yes", "yes"), SNP = c("no", "yes", "no", "no", "no", "no", 
    "no", "no", "no", "no", "no", "yes", "no")), .Names = c("typestudy", 
"dloop", "cytb", "coi", "other", "microsat", "SNP"), class = "data.frame", row.names = c(NA, 
-13L))