Question

我目前有一个“长”数据框，看起来像这样：

# Phylum          Sample  Abundance
# Other           A1      17
# Other           A2      0
# Actinobacteria  A3      123
# Actinobacteria  B1      651
# Other           B2      22
# Actinobacteria  B3      844
# Firmicutes      A1      403
# Other           A2      42
# Firmicutes      A3      225
# Other           B1      16
# Other           B2      19
# Firmicutes      B3      556

我的目标是使它看起来像这样，在该数据中，数据被旋转到较宽的位置，但是对重复的值求和（例如Other / A2和Other / B2）：

# Phylum          A1    A2    A3    B1    B2    B3
# Actinobacteria  0     0     123   651   0     844
# Firmicutes      403   0     225   0     0     556
# Other           17    71    0     16    41    0

到目前为止，我已经尝试过

newdata <- olddata %>% pivot_wider(names_from="Sample",values_from="Abundance",values_fill=0,values_fn=sum)

但这给出了错误Error in values_fn[[value]] : object of type 'builtin' is not subsettable。我曾尝试在本网站的其他地方查找，但大多数人建议制作唯一的行名，这对我而言不是一个选择。任何帮助将不胜感激。谢谢！

Answer 1

无法重现该错误，但最好将values_fill和values_fn放在list

中

library(tidyr)
library(dplyr)
olddata %>%        
    pivot_wider(names_from= Sample ,values_from= Abundance,
       values_fill=list(Abundance = 0),values_fn= list(Abundance = sum))
# A tibble: 3 x 7
#  Phylum            A1    A2    A3    B1    B2    B3
#  <chr>          <int> <int> <int> <int> <int> <int>
#1 Other             17    42     0    16    41     0
#2 Actinobacteria     0     0   123   651     0   844
#3 Firmicutes       403     0   225     0     0   556

数据

olddata <- structure(list(Phylum = c("Other", "Other", "Actinobacteria", 
"Actinobacteria", "Other", "Actinobacteria", "Firmicutes", "Other", 
"Firmicutes", "Other", "Other", "Firmicutes"), Sample = c("A1", 
"A2", "A3", "B1", "B2", "B3", "A1", "A2", "A3", "B1", "B2", "B3"
), Abundance = c(17L, 0L, 123L, 651L, 22L, 844L, 403L, 42L, 225L, 
16L, 19L, 556L)), class = "data.frame", row.names = c(NA, -12L
))

用pivot_wider（）对重复的列求和

1 个答案:

数据