如何在 R 的 for 循环中包含 left_join？

Question

我有几个具有丰富特定功能（OTU ID）的数据框。 以table1为例：

A tibble: 6 x 46
  `OTU ID` V1.1.K1 V1.1.K2 V1.1.K3 V1.1.N1 V1.1.N2 V1.14.K16 V1.14.K17 V1.14.K18 V1.14.N0 V1.14.W16 V1.14.W17
  <chr>      <dbl>   <dbl>   <dbl>   <dbl>   <dbl>     <dbl>     <dbl>     <dbl>    <dbl>     <dbl>     <dbl>
1 107273e~    2239    2861    1404   10255     585      4448      8340      5027      934       241      4983
2 bf83e8b~     174      95     406    4054     408     23383      6820     31184     2046       411       645
3 b744eae~      49      35     126    3440     245        83       443        18    11205       121       388
4 0e40fa4~      86      38     165    1015     111     10022      9973       228       90       638      2671
5 33ee4bb~      50      50      53      68      34       665       278      4816      218        38       136
6 c1c17b3~    2366    7838    7827     471      60       160        27        37      190        12       303
# ... with 34 more variables: V1.14.W18 <dbl>, V1.2.K4 <dbl>, V1.2.K5 <dbl>, V1.2.K6 <dbl>, V1.2.W4 <dbl>,
#   V1.2.W5 <dbl>, V1.2.W6 <dbl>, V1.21.K19 <dbl>, V1.21.K20 <dbl>, V1.21.K21 <dbl>, V1.21.N0 <dbl>,
#   V1.21.W19 <dbl>, V1.21.W20 <dbl>, V1.21.W21 <dbl>, V1.3.K7 <dbl>, V1.3.K8 <dbl>, V1.3.K9 <dbl>,
#   V1.3.W7 <dbl>, V1.3.W8 <dbl>, V1.3.W9 <dbl>, V1.4.K10 <dbl>, V1.4.K11 <dbl>, V1.4.K12 <dbl>, V1.4.W10 <dbl>,
#   V1.4.W11 <dbl>, V1.4.W12 <dbl>, V1.7.K13 <dbl>, V1.7.K14 <dbl>, V1.7.K15 <dbl>, V1.7.N0 <dbl>,
#   V1.7.W13 <dbl>, V1.7.W14 <dbl>, V1.7.W15 <dbl>, sum <dbl>

其他表看起来可比，它们只有丰度不同。然后我有一个分类数据框，其中包含有关特征的信息（OTU ID）。它被称为分类，看起来像这样：

`OTU ID`                 Taxon                                                                       Confidence
  <chr>                    <chr>                                                                            <dbl>
1 00009522ba8d1badfe92207~ d__Bacteria; p__Desulfobacterota; c__Desulfovibrionia; o__Desulfovibrional~      0.966
2 0000fb10633e8ad10a0f0b7~ d__Bacteria; p__Actinobacteriota; c__Actinobacteria; o__Micrococcales; f__~      0.999
3 0001d123420b59585627edf~ d__Bacteria; p__Proteobacteria; c__Alphaproteobacteria; o__Rhizobiales; f_~      1.00 
4 0001f7b8d2a9d04fbd0df3e~ d__Bacteria; p__Bacteroidota; c__Bacteroidia; o__Cytophagales; f__Bernarde~      1.00 
5 00026915bdc887a8587e62e~ d__Bacteria; p__Planctomycetota; c__Pla4_lineage; o__Pla4_lineage; f__Pla4~      0.997
6 0004a49903f29d739411b1c~ d__Archaea; p__Nanoarchaeota; c__Nanoarchaeia; o__Woesearchaeales; f__GW20~      0.777

由于 OTU ID - 列，丰度表应与分类表合并。 对于一个数据框，它使用以下代码：

table1_tax <- left_join(table1, taxonomy, by = "OTU ID")

现在我想将命令包含在一个 for 循环中，以循环我所有的丰度表， 每一个都应该与分类文件结合起来，并分配给一个我可以处理的新变量。 我不擅长 for 循环，请原谅我的错误。

#' get the tables in a list
tables <- mget(ls()[1:9])

#' as left_join can not loop over a list, I just extracted the names of the abundance tables in the variable. Was this step correct?
tables <- names(tables)

#' For Loop
for(i in 1:length(tables)) {
  assign(paste0("tax_", tables[i]), left_join(tables[i], taxonomy, by= setnames("OTU ID")))
}

我得到的错误是：

Error in UseMethod("left_join") : 
  no applicable method for 'left_join' applied to an object of class "character"

我该如何解决这个问题？你有什么主意吗？ 谢谢 ：） 凯瑟琳

Answer 1

以下内容能满足您的需求吗？

library(tidyverse)

# fake tables
df1 <- tribble(
  ~OTUID, ~V1, ~V2, ~V3, ~V4,
  "09522ba", 2239, 2861, 1404, 10255,
  "0fb1063", 174, 95, 406, 4054,
  "b744eae", 49, 35, 126, 3440
)

df2 <- df3 <- df4 <- df1

# named list of tables
lst_tables <- list(df1 = df1, df2 = df2, df3 = df3, df4 = df4)

# fake taxonomy
taxo <-
  tribble(
    ~OTUID, ~Taxon, ~Confidence,
    "09522ba", "d__Bacteria p__Desulfobacterota c__Desulfovibrionia", 0.966,
    "0fb1063", "d__Bacteria p__Actinobacteriota c__Actinobacteria", 0.999,
    "1d12342", "d__Bacteria p__Proteobacteria c__Alphaproteobacteria", 1.00,
  )

# 1) if you really want the loop
outbis <- list()

for(i in seq_along(lst_tables)) {
  outbis[[i]] <- left_join(lst_tables[[i]], taxo, by = "OTUID")
  names(outbis)[i] <- paste0("tax_", names(lst_tables)[i])
}

# 2) without the loop
out <-
  map(lst_tables, ~ left_join(.x, taxo))