Question

我有数据框我想找到具有相应值的公共列名

Project

输出

202055_at s100.Probeset，s101.Probeset，s102.Probeset，s103.Probeset，s10.Probeset

219957_at s102.Probeset，s103.Probeset，s10.Probeset

203063_at s100.Probeset，s10.Probeset

Answer 1

我们可以从数据集中获取unique个元素

un1 <- as.character(unique(unlist(df1)))

检查是否在多于1列中找到它，然后paste

out <- unlist(lapply(un1, function(x)  {
        x1 <- colSums(df1 == x)
    if(sum(x1) > 1)  paste(x, paste(names(x1)[x1>0], collapse="."))}))
out[1:2]
#[1] "202055_at s100.Probeset.s101.Probeset.s102.Probeset.s10.Probeset" 
#[2] "203063_at s100.Probeset.s10.Probeset"

Answer 2

只是为了扩展@ akrun的答案。

我不确定你要做什么，但是二进制矩阵表示哪个探针组在哪个列中可能对进一步处理有用。

# Your sample data
probes <- structure(list(s100.Probeset = structure(c(1L, 2L, 4L, 3L, 5L
 ), .Label = c("202055_at", "203063_at", "206284_x_at", "210734_x_at",
 "221915_s_at"), class = "factor"), s101.Probeset = structure(c(2L,
 4L, 3L, 1L, 5L), .Label = c("202055_at", "203248_at ", "206284_x_at",
 "210734_x_at", "212522_at"), class = "factor"), s102.Probeset = structure(c(1L,
 3L, 4L, 5L, 2L), .Label = c("202055_at", "205453_at", "210734_x_at",
"219957_at", "220661_s_at"), class = "factor"), s10.Probeset =  structure(c(5L,
1L, 2L, 3L, 4L), .Label = c("202055_at", "203063_at", "211503_s_at",
"214689_at", "219957_at"), class = "factor")), .Names = c("s100.Probeset",
"s101.Probeset", "s102.Probeset", "s10.Probeset"), row.names = c(NA,
-5L), class = "data.frame")

# Unique probesets
unique.probes <- unique(unlist(probes));

# Binary matrix
df1 <- data.frame(
    apply(probes, 2, function(x) unique.probes %in% x),
    row.names = unique.probes);
df1;
#            s100.Probeset s101.Probeset s102.Probeset s10.Probeset
#202055_at            TRUE          TRUE          TRUE         TRUE
#203063_at            TRUE         FALSE         FALSE         TRUE
#210734_x_at          TRUE          TRUE          TRUE        FALSE
#206284_x_at          TRUE          TRUE         FALSE        FALSE
#221915_s_at          TRUE         FALSE         FALSE        FALSE
#203248_at           FALSE          TRUE         FALSE        FALSE
#212522_at           FALSE          TRUE         FALSE        FALSE
#219957_at           FALSE         FALSE          TRUE         TRUE
#220661_s_at         FALSE         FALSE          TRUE        FALSE
#205453_at           FALSE         FALSE          TRUE        FALSE
#211503_s_at         FALSE         FALSE         FALSE         TRUE
#214689_at           FALSE         FALSE         FALSE         TRUE

然后，您可以将圈子关闭到@ akrun的解决方案。

# IDs per probeset
df2 <- cbind.data.frame(
    probeset = rownames(df1),
    ID = sapply(1:nrow(df1), function(x)
        paste(colnames(df1)[which(df1[x, ] == 1)], collapse = ",")));
df2;
#      probeset                                                   ID
#1    202055_at s100.Probeset,s101.Probeset,s102.Probeset,s10.Probeset
#2    203063_at                             s100.Probeset,s10.Probeset
#3  210734_x_at              s100.Probeset,s101.Probeset,s102.Probeset
#4  206284_x_at                            s100.Probeset,s101.Probeset
#5  221915_s_at                                          s100.Probeset
#6   203248_at                                           s101.Probeset
#7    212522_at                                          s101.Probeset
#8    219957_at                             s102.Probeset,s10.Probeset
#9  220661_s_at                                          s102.Probeset
#10   205453_at                                          s102.Probeset
#11 211503_s_at                                           s10.Probeset
#12   214689_at                                           s10.Probeset

如何在R中提取具有公共值的列名

2 个答案: