如何根據其他 3 列 [R] 中的匹配值計算我在 1 列中求和的行數？

Question

我有一張看起來像這樣的表：

> dt
                      variant_id           transcript_id
       1:  chr1_37492738_T_C_b38  chr1_37557076_37557602
       2:  chr1_37492738_T_C_b38  chr1_37557076_37557602
       3:  chr1_37492738_T_C_b38  chr1_37557076_37557602
       4:  chr1_37492738_T_C_b38  chr1_37557076_37557602
       5:  chr1_37492738_T_C_b38  chr1_37557076_37557602
      ---
13527497: chr22_49950090_T_G_b38 chr22_49925558_49927254
13527498: chr22_49950090_T_G_b38 chr22_49925558_49927254
13527499: chr22_49950090_T_G_b38 chr22_49925558_49927254
13527500: chr22_49950090_T_G_b38 chr22_49925558_49927254
13527501: chr22_49950090_T_G_b38 chr22_49925558_49927254
                         tissue_id counts individual is_NL
       1: GTEX-11DXX-1426-SM-5GIDU     46 GTEX-11DXX     0
       2: GTEX-11EM3-1726-SM-5N9D1     54 GTEX-11EM3     0
       3: GTEX-11EMC-1726-SM-5H11P     61 GTEX-11EMC     0
       4: GTEX-11GSP-0226-SM-5A5KV     44 GTEX-11GSP     0
       5: GTEX-11I78-1926-SM-59878     27 GTEX-11I78     0
      ---
13527497:  GTEX-ZVT2-0326-SM-5E44G    110  GTEX-ZVT2     1
13527498:  GTEX-ZVT3-2626-SM-5GU5L     54  GTEX-ZVT3     1
13527499:  GTEX-ZYFG-1726-SM-5GZZB     66  GTEX-ZYFG     1
13527500:  GTEX-ZYY3-2726-SM-5EGH4     96  GTEX-ZYY3     1
13527501:  GTEX-ZZPU-2126-SM-5EGIU     75  GTEX-ZZPU     0

通過使用以下行，我成功地總結了dt$counts的值： dt2 <- as.data.table(ddply(dt, c("variant_id", "transcript_id", "is_NL"), numcolwise(sum)))

使結果看起來像這樣：

> dt2
                     variant_id             transcript_id is_NL counts
     1: chr10_125381862_C_T_b38 chr10_124989699_124992694     0  30610
     2: chr10_125381862_C_T_b38 chr10_124989699_124992694     1   1932
     3: chr10_125381862_C_T_b38 chr10_124992813_124993201     0  28215
     4: chr10_125381862_C_T_b38 chr10_124992813_124993201     1   1706
     5: chr10_125381862_C_T_b38 chr10_124993330_124993854     0  17974
    ---
232637:   chr9_92815645_A_C_b38    chr9_92517876_92522574     1   2009
232638:   chr9_92815645_A_C_b38    chr9_92522894_92535932     0  10026
232639:   chr9_92815645_A_C_b38    chr9_92522894_92535932     1   1454
232640:   chr9_92815645_A_C_b38    chr9_92535983_92536600     0   2495
232641:   chr9_92815645_A_C_b38    chr9_92535983_92536600     1    341

但是，我還想要右側的一列，該列還指示對dt2$counts的值求和的行dt2$counts ，即每組匹配c("variant_id", "transcript_id", "is_NL")的樣本數. 我該怎么做呢？ 我可以弄清楚它是 Python 還是 Java，但不幸的是，對於 R，我什至不知道從哪里開始。

Answer 1

使用 dplyr：

library(dplyr)
dt2<- dt %>%
      group_by(variant_id, transcript_id, is_nl) %>%
      summarise(counts=sum(counts), nrows=n())

Answer 2

您也可以只使用不需要任何庫的聚合函數：

dt2 <- aggregate(counts ~ variant_id+transcript_id+is_NL, dt, function(x) c(sum = sum(x), n = length(x)))

Answer 3

在data.table ，它是：

setDT(dt)[, .(counts=sum(counts), .N), .(variant_id, transcript_id, is_NL)]