输入的每一行仅保留 unique/distinct 列 table
Retain only unique/distinct columns for each row of an input table
我有一个非常大的数据框 (nrow=~273,000)，我将其子集作为下面的示例：每一行都是一个蛋白质名称，并且有不同数量的列，列出了其中的亚细胞结构它们存在于人体细胞中。 1）我想删除每一行的重复条目并且正在努力解决这个问题（下面的代码）。 2) 然后我希望能够计算出每个基因可以在多少列（亚细胞结构）中找到。
背景：我从 Uniprot 获得了这些数据，并尽可能使用正则表达式对其进行了清理，但在某些情况下仍然存在具有重复条目的行（例如 FMR1 列出了染色体 2x、细胞质 3x 和质膜 2x - 此外，它们之间还有一些空白列）
dput(df1)
structure(list(FMR1 = structure(c(41L, 3L, 17L, 63L, 16L, 24L,
35L, 33L, 52L, 6L, 49L, 5L, 71L, 72L, 42L, 58L, 22L, 20L, 19L,
80L, 9L, 51L, 66L, 64L, 23L, 14L, 60L, 45L, 28L, 54L, 7L, 30L,
29L, 44L, 53L, 8L, 69L, 79L, 10L, 11L, 26L, 37L, 39L, 40L, 82L,
73L, 18L, 21L, 27L, 47L, 4L, 46L, 1L, 13L, 36L, 70L, 74L, 67L,
78L, 77L, 61L, 62L, 31L, 56L, 34L, 57L, 25L, 81L, 75L, 59L, 2L,
65L, 55L, 38L, 50L, 68L, 32L, 12L, 43L, 15L, 48L, 76L), .Label = c("AAMP",
"ADCY10 SAC", "AIMP1 EMAP2 SCYE1", "ANTXR2 CMG2", "APBB1 FE65 RIR",
"APC DP2", "APLP1", "ARHGAP26 GRAF KIAA0621 OPHN1L", "ARL4A ARL4",
"ATP6V0D1 ATP6D VPATPD", "ATP6V1D ATP6M VATD", "AZIN2 ADC KIAA1945 ODCP",
"CACNB2 CACNLB2 MYSB", "CAMK2D CAMKD", "CDCA8 PESCRG3", "CDK1 CDC2 CDC28A CDKN1 P34CDC2",
"CEMIP KIAA1199", "CIB1 CIB KIP PRKDCIP", "CLTA", "CLTB", "CMTM8 CKLFSF8",
"DMD", "DSP", "ECT2", "EHD2 PAST2", "ENTPD2 CD39L1", "ERBB2 HER2 MLN19 NEU NGL",
"EVPL", "FCHO1 KIAA0290", "FCHO2", "FGR SRC2", "GPER1 CEPR CMKRL2 DRY12 GPER GPR30",
"HDAC6 KIAA0901 JM21", "ITCH", "ITGB1BP1 ICAP1", "KCTD7", "KIFC3",
"MFN1", "MISP C19orf21", "MYOT TTID", "NGDN C14orf120", "NISCH IRAS KIAA0975",
"NR1D1 EAR1 HREV THRAL", "PGM5 PGMRP", "PKP4", "PLA2G6 PLPLA9",
"PNKD KIAA1184 MR1 TAHCCP2 FKSG19 UNQ2491/PRO5778", "POP7 RPP20",
"PPL KIAA0568", "PRDX3 AOP1", "PTOV1 ACID2 PP642 UNQ6127/PRO20092",
"PTPN23 KIAA1471", "PTPRE", "PTPRR ECPTP PTPRQ", "RAB13 GIG4",
"RAB23 HSPC137", "RAB29 RAB7L1", "RAB30", "RAB38", "RAB40AL RLGP",
"RAB8A MEL RAB8", "RAB9A RAB9", "RACGAP1 KIAA1478 MGCRACGAP",
"RAP1B OK/SW-cl", "RGS8", "RPSA LAMBR LAMR1", "SGIP1", "SHMT2",
"SHROOM3 KIAA1481 SHRML MSTP013", "SLC28A3 CNT3", "SNTA1 SNT1",
"SNTB1 SNT2B1", "SNX11", "SNX12", "STOM BND7 EPB72", "TEX10 L18 Nbla10363",
"TNFRSF8 CD30 D1S166E", "TNS4 CTEN PP14434", "TRIM72 MG53", "USP6 HRP1 TRE2",
"VCL", "YES1 YES"), class = "factor"), Nucleus = structure(c(3L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 5L,
5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L,
2L), .Label = c("Mitochondrion ", "Nucleus", "Nucleus ", "Plasma membrane",
"Plasma membrane "), class = "factor"), Chromosome = structure(c(1L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L,
3L), .Label = c("Chromosome", "Cytoplasm", "Cytoplasm "), class = "factor"),
Chromosome.1 = structure(c(4L, 5L, 7L, 5L, 14L, 12L, 20L,
18L, 5L, 20L, 20L, 2L, 1L, 1L, 8L, 10L, 19L, 1L, 1L, 8L,
16L, 16L, 17L, 19L, 20L, 21L, 15L, 13L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 9L, 10L,
16L, 16L, 16L, 22L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 11L,
7L, 14L, 9L, 17L, 11L, 9L, 2L, 6L, 6L, 17L, 18L, 10L, 1L,
1L, 17L, 19L, 19L, 1L, 3L, 5L, 1L), .Label = c("", " ", "Chromosome",
"Cytoplasm ", "Cytoplasmic vesicle", "Cytoplasmic vesicle ",
"Endoplasmic reticulum", "Endosome", "Endosome ", "Golgi apparatus",
"Golgi apparatus ", "Midbody", "Midbody ", "Mitochondrion",
"Mitochondrion ", "Nucleus", "Nucleus ", "Perikaryon ",
"Plasma membrane", "Plasma membrane ", "Sarcoplasmic reticulum ",
"Secreted"), class = "factor"), Cytoplasm = structure(c(1L,
15L, 13L, 10L, 1L, 13L, 1L, 1L, 5L, 2L, 11L, 1L, 1L, 1L,
5L, 8L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 14L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 5L, 9L, 2L, 3L, 6L, 7L, 2L, 1L, 2L, 4L, 11L, 12L,
5L, 1L, 1L, 1L, 7L, 3L, 1L, 2L, 2L, 2L), .Label = c("", " ",
"Cytoplasmic vesicle", "Endoplasmic reticulum", "Endosome",
"Endosome ", "Golgi apparatus", "Golgi apparatus ", "Golgi appartus",
"Midbody", "Mitochondrion ", "Nucleus ", "Plasma membrane",
"Plasma membrane ", "Secreted "), class = "factor"), Cytoplasm.1 = structure(c(1L,
4L, 7L, 7L, 1L, 1L, 1L, 1L, 5L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
6L, 3L, 2L, 1L, 1L, 1L), .Label = c("", " ", "Endoplasmic reticulum",
"Endoplasmic reticulum ", "Endosome", "Mitochondrion", "Plasma membrane"
), class = "factor"), Cytoplasmic.vesicle = structure(c(1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 1L, 1L), .Label = c("", "Golgi apparatus"
), class = "factor"), Perikaryon = structure(c(2L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L,
1L, 1L, 1L, 1L), .Label = c("", " ", "Golgi apparatus"), class = "factor"),
X = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L), .Label = c("",
"Cytoplasmic granule"), class = "factor"), X.1 = structure(c(1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 1L, 1L), .Label = c("", "Perikaryon"), class = "factor"),
X.2 = c(NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA), X.3 = c(NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA), Plasma.membrane = c(NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA), Plasma.membrane.1 = c(NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
)), .Names = c("FMR1", "Nucleus", "Chromosome", "Chromosome.1",
"Cytoplasm", "Cytoplasm.1", "Cytoplasmic.vesicle", "Perikaryon",
"X", "X.1", "X.2", "X.3", "Plasma.membrane", "Plasma.membrane.1"
), class = "data.frame", row.names = c(NA, -82L))
我试过只为每一行获取唯一的列，但没有成功，例如：
unique(df1) # Original data with repeats removed
dplyr::distinct(df1) # Retain only unique/distinct rows from an input tb
我认为问题在于上述函数正在寻找相同的行名称，这不是我想要的。我希望每一行都有不同的列。我正在考虑使用 melt 函数，但由于每行的列数为奇数，所以这行不通。
我希望输出看起来像这样 newDF
structure(list(FMR1 = structure(c(7L, 1L, 3L, 9L, 2L, 4L, 6L,
5L, 8L), .Label = c("AIMP1 EMAP2 SCYE1", "CDK1 CDC2 CDC28A CDKN1 P34CDC2",
"CEMIP KIAA1199", "ECT2", "HDAC6 KIAA0901 JM21", "ITGB1BP1 ICAP1",
"NGDN C14orf120", "PTPN23 KIAA1471", "RACGAP1 KIAA1478 MGCRACGAP"
), class = "factor"), Nucleus = structure(c(2L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L), .Label = c("Nucleus", "Nucleus "), class = "factor"),
Chromosome = structure(c(1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L), .Label = c("Chromosome", "Cytoplasm"), class = "factor"),
Cytoplasmic.vesicle = structure(c(1L, 8L, 2L, 4L, 5L, 4L,
7L, 6L, 3L), .Label = c("Cytoplasm ", "Endoplasmic reticulum",
"Endosome", "Midbody", "Mitochondrion", "Perikaryon ", "Plasma membrane ",
"Secreted "), class = "factor"), Perikaryon = structure(c(1L,
2L, 3L, 3L, 1L, 3L, 1L, 1L, 1L), .Label = c("", "Endoplasmic reticulum ",
"Plasma membrane"), class = "factor"), Plasma.membrane = c(NA,
NA, NA, NA, NA, NA, NA, NA, NA)), .Names = c("FMR1", "Nucleus",
"Chromosome", "Cytoplasmic.vesicle", "Perikaryon", "Plasma.membrane"
), class = "data.frame", row.names = c(NA, -9L))
从这里我想得到一个 rowSums(df1) 所以我正在考虑将每个术语强制转换为一个数字（例如细胞质囊泡=1，细胞核=1，内质网=1等）但是运行 进入这个虚拟数据集的问题。
df2 <- as.numeric(newDF)
Error: (list) object cannot be coerced to type 'double'
df2 <- as.numeric(newDF[,2:n])
Error in 2:n : NA/NaN argument
感谢您的帮助。
编辑
我想计算 newDF 中每一行的唯一列数，如下所示：
FMR1 5
NGDN C14orf120 3
AIMP1 EMAP2 SCYE1 4
CEMIP KIAA1199 4
RACGAP1 KIAA1478 MGCRACGAP 4
CDK1 CDC2 CDC28A CDKN1 P34CDC2 3
ECT2 4
ITGB1BP1 ICAP1 3
HDAC6 KIAA0901 JM21 3
PTPN23 KIAA1471 3
这可能是一种方法。由于您的预期结果是字符向量，因此我无法将最终输出可视化。然而，您说您想检查每种蛋白质在数据中出现的列数。希望我的结果是你想要的。
首先，我将所有列都转换为字符。然后，我使用 gather() 将数据转换为长格式。对于每个亚细胞结构组（即亚细胞），我添加了行索引（例如，1 表示原始数据中的第一行）和 trim 白色 space。然后，删除蛋白质中带有 NA 的所有行。删除任何包含 "" 和 " " 的行。现在整理工作完成了。对于每一行（即 row.index），删除重复的蛋白质类型。
取消分组数据，最后计算每种蛋白质出现的列数（即，细胞结构）。基本上，你想统计此时每种蛋白质在数据集中出现了多少次。
根据您的示例数据，我得到了以下结果。但我不确定这是否是你想要的。 （我现在要去睡觉了。所以我有几个小时无法帮助你。如果有人可以加入，请加入。）
mutate_all(mydf, as.character) %>%
gather(key = subcellular, value = protein) %>%
group_by(subcellular) %>%
mutate(row.index = 1:n(),
protein = trimws(protein)) %>%
filter(!is.na(protein)) %>%
filter(!protein %in% c("", " ")) %>%
group_by(row.index) %>%
filter(!duplicated(protein)) %>%
ungroup %>%
count(protein, sort = TRUE)
# protein n
# <chr> <int>
# 1 Cytoplasm 82
# 2 Plasma membrane 70
# 3 Nucleus 25
# 4 Endosome 9
# 5 Mitochondrion 9
# 6 Cytoplasmic vesicle 8
# 7 Golgi apparatus 7
# 8 Endoplasmic reticulum 5
# 9 Midbody 3
#10 Perikaryon 3
# ... with 87 more rows
根据 jjl 的评论，我做了以下操作。我没有计算每个蛋白质出现的列数，而是计算每一行存在多少个蛋白质名称。
mutate_all(mydf, as.character) %>%
gather(key = subcellular, value = protein) %>%
group_by(subcellular) %>%
mutate(row.index = 1:n(),
protein = trimws(protein)) %>%
filter(!is.na(protein)) %>%
filter(!protein %in% c("", " ")) %>%
group_by(row.index) %>%
filter(!duplicated(protein)) %>%
ungroup %>%
count(row.index)
# row.index n
# <int> <int>
# 1 1 4
# 2 2 6
# 3 3 5
# 4 4 6
# 5 5 4
# 6 6 5
# 7 7 4
# 8 8 4
# 9 9 5
#10 10 3
# ... with 72 more rows
编辑
如果您想删除第一列（即 FMR1），您可以通过过滤该列来实现。在最后使用 count() 之前，我在代码中添加了 filter(subcellular != "FMR1")。
mutate_all(mydf, as.character) %>%
gather(key = subcellular, value = protein) %>%
group_by(subcellular) %>%
mutate(row.index = 1:n(),
protein = trimws(protein)) %>%
filter(!is.na(protein)) %>%
filter(!protein %in% c("", " ")) %>%
group_by(row.index) %>%
filter(!duplicated(protein)) %>%
ungroup %>%
filter(subcellular != "FMR1") %>%
count(row.index)
# A tibble: 9 x 2
# row.index n
# <int> <int>
#1 1 3
#2 2 4
#3 3 4
#4 4 4
#5 5 3
#6 6 4
#7 7 3
#8 8 3
#9 9 3
我有一个非常大的数据框 (nrow=~273,000)，我将其子集作为下面的示例：每一行都是一个蛋白质名称，并且有不同数量的列，列出了其中的亚细胞结构它们存在于人体细胞中。 1）我想删除每一行的重复条目并且正在努力解决这个问题（下面的代码）。 2) 然后我希望能够计算出每个基因可以在多少列（亚细胞结构）中找到。
背景：我从 Uniprot 获得了这些数据，并尽可能使用正则表达式对其进行了清理，但在某些情况下仍然存在具有重复条目的行（例如 FMR1 列出了染色体 2x、细胞质 3x 和质膜 2x - 此外，它们之间还有一些空白列）
dput(df1)
structure(list(FMR1 = structure(c(41L, 3L, 17L, 63L, 16L, 24L,
35L, 33L, 52L, 6L, 49L, 5L, 71L, 72L, 42L, 58L, 22L, 20L, 19L,
80L, 9L, 51L, 66L, 64L, 23L, 14L, 60L, 45L, 28L, 54L, 7L, 30L,
29L, 44L, 53L, 8L, 69L, 79L, 10L, 11L, 26L, 37L, 39L, 40L, 82L,
73L, 18L, 21L, 27L, 47L, 4L, 46L, 1L, 13L, 36L, 70L, 74L, 67L,
78L, 77L, 61L, 62L, 31L, 56L, 34L, 57L, 25L, 81L, 75L, 59L, 2L,
65L, 55L, 38L, 50L, 68L, 32L, 12L, 43L, 15L, 48L, 76L), .Label = c("AAMP",
"ADCY10 SAC", "AIMP1 EMAP2 SCYE1", "ANTXR2 CMG2", "APBB1 FE65 RIR",
"APC DP2", "APLP1", "ARHGAP26 GRAF KIAA0621 OPHN1L", "ARL4A ARL4",
"ATP6V0D1 ATP6D VPATPD", "ATP6V1D ATP6M VATD", "AZIN2 ADC KIAA1945 ODCP",
"CACNB2 CACNLB2 MYSB", "CAMK2D CAMKD", "CDCA8 PESCRG3", "CDK1 CDC2 CDC28A CDKN1 P34CDC2",
"CEMIP KIAA1199", "CIB1 CIB KIP PRKDCIP", "CLTA", "CLTB", "CMTM8 CKLFSF8",
"DMD", "DSP", "ECT2", "EHD2 PAST2", "ENTPD2 CD39L1", "ERBB2 HER2 MLN19 NEU NGL",
"EVPL", "FCHO1 KIAA0290", "FCHO2", "FGR SRC2", "GPER1 CEPR CMKRL2 DRY12 GPER GPR30",
"HDAC6 KIAA0901 JM21", "ITCH", "ITGB1BP1 ICAP1", "KCTD7", "KIFC3",
"MFN1", "MISP C19orf21", "MYOT TTID", "NGDN C14orf120", "NISCH IRAS KIAA0975",
"NR1D1 EAR1 HREV THRAL", "PGM5 PGMRP", "PKP4", "PLA2G6 PLPLA9",
"PNKD KIAA1184 MR1 TAHCCP2 FKSG19 UNQ2491/PRO5778", "POP7 RPP20",
"PPL KIAA0568", "PRDX3 AOP1", "PTOV1 ACID2 PP642 UNQ6127/PRO20092",
"PTPN23 KIAA1471", "PTPRE", "PTPRR ECPTP PTPRQ", "RAB13 GIG4",
"RAB23 HSPC137", "RAB29 RAB7L1", "RAB30", "RAB38", "RAB40AL RLGP",
"RAB8A MEL RAB8", "RAB9A RAB9", "RACGAP1 KIAA1478 MGCRACGAP",
"RAP1B OK/SW-cl", "RGS8", "RPSA LAMBR LAMR1", "SGIP1", "SHMT2",
"SHROOM3 KIAA1481 SHRML MSTP013", "SLC28A3 CNT3", "SNTA1 SNT1",
"SNTB1 SNT2B1", "SNX11", "SNX12", "STOM BND7 EPB72", "TEX10 L18 Nbla10363",
"TNFRSF8 CD30 D1S166E", "TNS4 CTEN PP14434", "TRIM72 MG53", "USP6 HRP1 TRE2",
"VCL", "YES1 YES"), class = "factor"), Nucleus = structure(c(3L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L,
4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 4L, 5L, 5L, 5L, 5L, 5L,
5L, 5L, 5L, 5L, 5L, 5L, 5L, 5L, 1L, 1L, 1L, 2L, 2L, 2L, 2L, 2L,
2L), .Label = c("Mitochondrion ", "Nucleus", "Nucleus ", "Plasma membrane",
"Plasma membrane "), class = "factor"), Chromosome = structure(c(1L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 3L, 3L, 3L, 3L, 3L, 3L, 3L, 3L,
3L), .Label = c("Chromosome", "Cytoplasm", "Cytoplasm "), class = "factor"),
Chromosome.1 = structure(c(4L, 5L, 7L, 5L, 14L, 12L, 20L,
18L, 5L, 20L, 20L, 2L, 1L, 1L, 8L, 10L, 19L, 1L, 1L, 8L,
16L, 16L, 17L, 19L, 20L, 21L, 15L, 13L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 2L, 2L, 2L, 9L, 10L,
16L, 16L, 16L, 22L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 11L,
7L, 14L, 9L, 17L, 11L, 9L, 2L, 6L, 6L, 17L, 18L, 10L, 1L,
1L, 17L, 19L, 19L, 1L, 3L, 5L, 1L), .Label = c("", " ", "Chromosome",
"Cytoplasm ", "Cytoplasmic vesicle", "Cytoplasmic vesicle ",
"Endoplasmic reticulum", "Endosome", "Endosome ", "Golgi apparatus",
"Golgi apparatus ", "Midbody", "Midbody ", "Mitochondrion",
"Mitochondrion ", "Nucleus", "Nucleus ", "Perikaryon ",
"Plasma membrane", "Plasma membrane ", "Sarcoplasmic reticulum ",
"Secreted"), class = "factor"), Cytoplasm = structure(c(1L,
15L, 13L, 10L, 1L, 13L, 1L, 1L, 5L, 2L, 11L, 1L, 1L, 1L,
5L, 8L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 2L, 1L, 1L, 14L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 5L, 9L, 2L, 3L, 6L, 7L, 2L, 1L, 2L, 4L, 11L, 12L,
5L, 1L, 1L, 1L, 7L, 3L, 1L, 2L, 2L, 2L), .Label = c("", " ",
"Cytoplasmic vesicle", "Endoplasmic reticulum", "Endosome",
"Endosome ", "Golgi apparatus", "Golgi apparatus ", "Golgi appartus",
"Midbody", "Mitochondrion ", "Nucleus ", "Plasma membrane",
"Plasma membrane ", "Secreted "), class = "factor"), Cytoplasm.1 = structure(c(1L,
4L, 7L, 7L, 1L, 1L, 1L, 1L, 5L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
6L, 3L, 2L, 1L, 1L, 1L), .Label = c("", " ", "Endoplasmic reticulum",
"Endoplasmic reticulum ", "Endosome", "Mitochondrion", "Plasma membrane"
), class = "factor"), Cytoplasmic.vesicle = structure(c(1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 1L, 1L), .Label = c("", "Golgi apparatus"
), class = "factor"), Perikaryon = structure(c(2L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 3L,
1L, 1L, 1L, 1L), .Label = c("", " ", "Golgi apparatus"), class = "factor"),
X = structure(c(1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 2L, 1L, 1L, 1L, 1L), .Label = c("",
"Cytoplasmic granule"), class = "factor"), X.1 = structure(c(1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L, 1L,
1L, 2L, 1L, 1L, 1L, 1L), .Label = c("", "Perikaryon"), class = "factor"),
X.2 = c(NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA), X.3 = c(NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA), Plasma.membrane = c(NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA), Plasma.membrane.1 = c(NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA,
NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA
)), .Names = c("FMR1", "Nucleus", "Chromosome", "Chromosome.1",
"Cytoplasm", "Cytoplasm.1", "Cytoplasmic.vesicle", "Perikaryon",
"X", "X.1", "X.2", "X.3", "Plasma.membrane", "Plasma.membrane.1"
), class = "data.frame", row.names = c(NA, -82L))
我试过只为每一行获取唯一的列，但没有成功，例如：
unique(df1) # Original data with repeats removed
dplyr::distinct(df1) # Retain only unique/distinct rows from an input tb
我认为问题在于上述函数正在寻找相同的行名称，这不是我想要的。我希望每一行都有不同的列。我正在考虑使用 melt 函数，但由于每行的列数为奇数，所以这行不通。
我希望输出看起来像这样 newDF
structure(list(FMR1 = structure(c(7L, 1L, 3L, 9L, 2L, 4L, 6L,
5L, 8L), .Label = c("AIMP1 EMAP2 SCYE1", "CDK1 CDC2 CDC28A CDKN1 P34CDC2",
"CEMIP KIAA1199", "ECT2", "HDAC6 KIAA0901 JM21", "ITGB1BP1 ICAP1",
"NGDN C14orf120", "PTPN23 KIAA1471", "RACGAP1 KIAA1478 MGCRACGAP"
), class = "factor"), Nucleus = structure(c(2L, 1L, 1L, 1L, 1L,
1L, 1L, 1L, 1L), .Label = c("Nucleus", "Nucleus "), class = "factor"),
Chromosome = structure(c(1L, 2L, 2L, 2L, 2L, 2L, 2L, 2L,
2L), .Label = c("Chromosome", "Cytoplasm"), class = "factor"),
Cytoplasmic.vesicle = structure(c(1L, 8L, 2L, 4L, 5L, 4L,
7L, 6L, 3L), .Label = c("Cytoplasm ", "Endoplasmic reticulum",
"Endosome", "Midbody", "Mitochondrion", "Perikaryon ", "Plasma membrane ",
"Secreted "), class = "factor"), Perikaryon = structure(c(1L,
2L, 3L, 3L, 1L, 3L, 1L, 1L, 1L), .Label = c("", "Endoplasmic reticulum ",
"Plasma membrane"), class = "factor"), Plasma.membrane = c(NA,
NA, NA, NA, NA, NA, NA, NA, NA)), .Names = c("FMR1", "Nucleus",
"Chromosome", "Cytoplasmic.vesicle", "Perikaryon", "Plasma.membrane"
), class = "data.frame", row.names = c(NA, -9L))
从这里我想得到一个 rowSums(df1) 所以我正在考虑将每个术语强制转换为一个数字（例如细胞质囊泡=1，细胞核=1，内质网=1等）但是运行 进入这个虚拟数据集的问题。
df2 <- as.numeric(newDF)
Error: (list) object cannot be coerced to type 'double'
df2 <- as.numeric(newDF[,2:n])
Error in 2:n : NA/NaN argument
感谢您的帮助。
编辑我想计算 newDF 中每一行的唯一列数，如下所示：
FMR1 5
NGDN C14orf120 3
AIMP1 EMAP2 SCYE1 4
CEMIP KIAA1199 4
RACGAP1 KIAA1478 MGCRACGAP 4
CDK1 CDC2 CDC28A CDKN1 P34CDC2 3
ECT2 4
ITGB1BP1 ICAP1 3
HDAC6 KIAA0901 JM21 3
PTPN23 KIAA1471 3
这可能是一种方法。由于您的预期结果是字符向量，因此我无法将最终输出可视化。然而，您说您想检查每种蛋白质在数据中出现的列数。希望我的结果是你想要的。
首先，我将所有列都转换为字符。然后，我使用 gather() 将数据转换为长格式。对于每个亚细胞结构组（即亚细胞），我添加了行索引（例如，1 表示原始数据中的第一行）和 trim 白色 space。然后，删除蛋白质中带有 NA 的所有行。删除任何包含 "" 和 " " 的行。现在整理工作完成了。对于每一行（即 row.index），删除重复的蛋白质类型。
取消分组数据，最后计算每种蛋白质出现的列数（即，细胞结构）。基本上，你想统计此时每种蛋白质在数据集中出现了多少次。
根据您的示例数据，我得到了以下结果。但我不确定这是否是你想要的。 （我现在要去睡觉了。所以我有几个小时无法帮助你。如果有人可以加入，请加入。）
mutate_all(mydf, as.character) %>%
gather(key = subcellular, value = protein) %>%
group_by(subcellular) %>%
mutate(row.index = 1:n(),
protein = trimws(protein)) %>%
filter(!is.na(protein)) %>%
filter(!protein %in% c("", " ")) %>%
group_by(row.index) %>%
filter(!duplicated(protein)) %>%
ungroup %>%
count(protein, sort = TRUE)
# protein n
# <chr> <int>
# 1 Cytoplasm 82
# 2 Plasma membrane 70
# 3 Nucleus 25
# 4 Endosome 9
# 5 Mitochondrion 9
# 6 Cytoplasmic vesicle 8
# 7 Golgi apparatus 7
# 8 Endoplasmic reticulum 5
# 9 Midbody 3
#10 Perikaryon 3
# ... with 87 more rows
根据 jjl 的评论，我做了以下操作。我没有计算每个蛋白质出现的列数，而是计算每一行存在多少个蛋白质名称。
mutate_all(mydf, as.character) %>%
gather(key = subcellular, value = protein) %>%
group_by(subcellular) %>%
mutate(row.index = 1:n(),
protein = trimws(protein)) %>%
filter(!is.na(protein)) %>%
filter(!protein %in% c("", " ")) %>%
group_by(row.index) %>%
filter(!duplicated(protein)) %>%
ungroup %>%
count(row.index)
# row.index n
# <int> <int>
# 1 1 4
# 2 2 6
# 3 3 5
# 4 4 6
# 5 5 4
# 6 6 5
# 7 7 4
# 8 8 4
# 9 9 5
#10 10 3
# ... with 72 more rows
编辑
如果您想删除第一列（即 FMR1），您可以通过过滤该列来实现。在最后使用 count() 之前，我在代码中添加了 filter(subcellular != "FMR1")。
mutate_all(mydf, as.character) %>%
gather(key = subcellular, value = protein) %>%
group_by(subcellular) %>%
mutate(row.index = 1:n(),
protein = trimws(protein)) %>%
filter(!is.na(protein)) %>%
filter(!protein %in% c("", " ")) %>%
group_by(row.index) %>%
filter(!duplicated(protein)) %>%
ungroup %>%
filter(subcellular != "FMR1") %>%
count(row.index)
# A tibble: 9 x 2
# row.index n
# <int> <int>
#1 1 3
#2 2 4
#3 3 4
#4 4 4
#5 5 3
#6 6 4
#7 7 3
#8 8 3
#9 9 3