Question

我想找出这两个表之间重叠的对：

> dput(data1)
structure(list(Name.x = c("MDH1", "MDH1", "IDH2", "IDH2", "IDH2", 
"IDH2", "IDH2", "IDH2", "IDH2", "SCOALB", "SCOALB", "CSY4", "CSY4", 
"CSY4", "CSY4", "CSY4", "FUM1", "FUM1", "IDH6", "IDH6", "IDH6", 
"ODC1-1", "ODC1-1", "ODC1-1", "ODC1-1", "ODC1-1", "ODC2-1", "ODC2-1", 
"ODC2-1", "ACO2", "IDH1", "IDH1", "IDH1", "IDH1", "ODC2-2"), 
    Name.y = c("SCOALB", "SCOALA-1", "CSY4", "IDH6", "ODC1-1", 
    "ODC2-1", "IDH1", "ODC2-2", "ODC1-2", "SCOALA-1", "SCOALA-2", 
    "IDH6", "SDH2-1", "IDH1", "IDH5", "ICDH", "ODC1-1", "ODC1-2", 
    "ACO2", "IDH1", "IDH5", "ODC2-1", "IDH1", "IDH5", "ODC2-2", 
    "ODC1-2", "IDH1", "ODC2-2", "ODC1-2", "IDH1", "IDH5", "SCOALA-2", 
    "ODC2-2", "ODC1-2", "ODC1-2")), .Names = c("Name.x", "Name.y"
), class = "data.frame", row.names = c(NA, -35L))



> dput(data2)
    structure(list(Protein1 = structure(c(3L, 7L, 18L, 19L, 7L, 19L, 
    6L, 18L, 6L, 18L, 18L, 19L, 9L, 8L, 19L, 18L, 9L, 7L, 18L, 12L, 
    8L, 19L, 5L, 29L, 12L, 29L, 12L, 18L, 7L, 17L, 6L, 5L, 9L, 19L, 
    12L, 3L, 19L, 16L, 18L, 17L, 16L, 17L, 9L, 29L, 12L, 7L, 29L, 
    18L, 16L, 18L, 29L, 8L, 17L, 16L, 17L, 12L, 6L, 8L, 17L, 29L, 
    9L, 17L, 29L, 19L, 8L, 17L, 29L, 9L, 9L, 16L, 29L, 29L, 19L, 
    19L, 19L, 29L, 12L, 19L, 17L, 29L, 17L, 16L, 16L, 19L, 16L, 4L, 
    1L, 5L, 17L, 9L, 18L, 18L, 6L, 4L, 8L, 16L, 16L, 29L, 7L, 12L, 
    8L, 4L, 29L, 12L, 5L), .Label = c("ACO2", "ACO3", "CSY4", "FUM1", 
    "ICDH", "IDH1", "IDH2", "IDH5", "IDH6", "LPD1", "LPD2", "MDH1", 
    "MDH2", "ME1", "ME2", "ODC1-1", "ODC1-2", "ODC2-1", "ODC2-2", 
    "PDC1a-1", "PDC1a-2", "PDC1b", "PDC2-1", "PDC2-2", "SCoALa-1", 
    "SCoALa-2", "SCoALb", "SDH1-1", "SDH2-1", "SDH2-2", "SDH2-3", 
    "SDH3-1", "SDH4", "SDH5", "SDH6", "SDH7a", "SDH7b", "SDH8"), class = "factor"), 
        Protein2 = structure(c(1L, 6L, 7L, 17L, 1L, 16L, 3L, 9L, 
        1L, 5L, 17L, 9L, 8L, 7L, 18L, 18L, 5L, 3L, 16L, 3L, 5L, 8L, 
        4L, 7L, 5L, 3L, 6L, 6L, 5L, 3L, 5L, 3L, 3L, 6L, 7L, 3L, 7L, 
        9L, 1L, 8L, 5L, 16L, 7L, 6L, 4L, 7L, 4L, 3L, 3L, 12L, 1L, 
        1L, 9L, 7L, 7L, 9L, 6L, 6L, 5L, 8L, 1L, 17L, 29L, 3L, 8L, 
        6L, 9L, 9L, 6L, 12L, 5L, 19L, 12L, 5L, 1L, 16L, 1L, 19L, 
        4L, 18L, 12L, 1L, 4L, 4L, 6L, 3L, 1L, 1L, 1L, 4L, 4L, 8L, 
        4L, 1L, 3L, 8L, 16L, 12L, 4L, 12L, 4L, 4L, 17L, 8L, 5L), .Label = c("ACO2", 
        "ACO3", "CSY4", "FUM1", "ICDH", "IDH1", "IDH2", "IDH5", "IDH6", 
        "LPD1", "LPD2", "MDH1", "MDH2", "ME1", "ME2", "ODC1-1", "ODC1-2", 
        "ODC2-1", "ODC2-2", "PDC1a-1", "PDC1a-2", "PDC1b", "PDC2-1", 
        "PDC2-2", "SCoALa-1", "SCoALa-2", "SCoALb", "SDH1-1", "SDH2-1", 
        "SDH2-2", "SDH2-3", "SDH3-1", "SDH4", "SDH5", "SDH6", "SDH7a", 
        "SDH7b", "SDH8"), class = "factor")), .Names = c("Protein1", 
    "Protein2"), class = "data.frame", row.names = c(1L, 4L, 6L, 
    12L, 22L, 25L, 28L, 33L, 44L, 48L, 51L, 52L, 53L, 60L, 68L, 70L, 
    72L, 76L, 86L, 109L, 110L, 119L, 133L, 144L, 146L, 158L, 170L, 
    197L, 202L, 206L, 211L, 213L, 226L, 227L, 237L, 271L, 272L, 286L, 
    290L, 297L, 304L, 305L, 306L, 319L, 323L, 327L, 347L, 348L, 351L, 
    357L, 370L, 372L, 373L, 378L, 379L, 392L, 406L, 410L, 414L, 417L, 
    419L, 437L, 442L, 445L, 448L, 455L, 457L, 462L, 471L, 479L, 482L, 
    483L, 488L, 503L, 509L, 522L, 536L, 563L, 618L, 620L, 623L, 628L, 
    630L, 644L, 647L, 666L, 668L, 673L, 676L, 678L, 679L, 690L, 691L, 
    694L, 698L, 703L, 709L, 714L, 715L, 722L, 723L, 724L, 727L, 739L, 
    740L))

在每个df中，有两列存储字符串。表之间的字符串重叠。但是，对之间的顺序可能不同。该对中的一个字符串可能会在data1的第一列和data2的第二列中找到。如何找到哪些对以及它们之间有多少重叠数据集？

Answer 1

> data1$combine = as.character(interaction(data1$Name.x, data1$Name.y))
> data2$combine = as.character(interaction(data2$Protein1, data2$Protein2))
> 
> dat.overlap = data1[complete.cases(match(data2$combine, data1$combine)),]
> dat.overlap
     Name.x   Name.y         combine
2      MDH1 SCOALA-1   MDH1.SCOALA-1
4      IDH2     IDH6       IDH2.IDH6
11   SCOALB SCOALA-2 SCOALB.SCOALA-2
13     CSY4   SDH2-1     CSY4.SDH2-1
18     FUM1   ODC1-2     FUM1.ODC1-2
28   ODC2-1   ODC2-2   ODC2-1.ODC2-2

data1[complete.cases(match(data1$combine, data2$combine)),]
   Name.x Name.y       combine
3    IDH2   CSY4     IDH2.CSY4
7    IDH2   IDH1     IDH2.IDH1
19   IDH6   ACO2     IDH6.ACO2
20   IDH6   IDH1     IDH6.IDH1
21   IDH6   IDH5     IDH6.IDH5
23 ODC1-1   IDH1   ODC1-1.IDH1
24 ODC1-1   IDH5   ODC1-1.IDH5
27 ODC2-1   IDH1   ODC2-1.IDH1
29 ODC2-1 ODC1-2 ODC2-1.ODC1-2
35 ODC2-2 ODC1-2 ODC2-2.ODC1-2

Answer 2

按行排序并通过粘贴创建一个键，然后合并：

data1$key <- apply(data1, 1, function(i) paste(sort(i), collapse = "_"))
data2$key <- apply(data2, 1, function(i) paste(sort(i), collapse = "_"))
res <- merge(data1, data2, by = "key")

head(res)
#         key Name.x Name.y Protein1 Protein2
# 1 ACO2_IDH1   ACO2   IDH1     IDH1     ACO2
# 2 ACO2_IDH6   IDH6   ACO2     IDH6     ACO2
# 3 CSY4_ICDH   CSY4   ICDH     ICDH     CSY4
# 4 CSY4_IDH1   CSY4   IDH1     IDH1     CSY4
# 5 CSY4_IDH2   IDH2   CSY4     IDH2     CSY4
# 6 CSY4_IDH5   CSY4   IDH5     IDH5     CSY4

从df中提取两列重叠的行

2 个答案: