时间戳

首页 > 解决方案 > For循环：如何为特定迭代应用不同的命令

r - For循环：如何为特定迭代应用不同的命令

问题描述

您好，我有以下内容data frame，我想对其进行分析，for loop但在分析代码中，我想仅删除一个基因（Hbvrt）的一个或多个样本，因此它们不应包含在整个分析中。

## create some data
sample_ID <- rep(c('s1','s2','s3','s4'),4)
gene_ID <- c( rep('TFT',4) , rep('Hbvrt' ,4), rep('Myx4',4), rep('Rai56n',4))
readz <- runif(16, 5000, 7500) 

df <- data.frame(sample_ID , gene_ID , readz)

## start the loop 
res <- list()
for ( g in unique(df$gene_ID)){
  
  df_g <- df[df$gene_ID == g, ]
  
  df_g$Nanost <- runif(4, 5000, 7500)
  
  df_g$NEW <- df_g$Nanost / df_g$readz * 100 

  ## AND long code here ....

## function for graph ## Graphs Not shown here  
  scatter_fun = function(x, y) {
    
    ggscatter(df_g, x = "readz", y = "Nanost", 
              add = "reg.line", conf.int = TRUE, 
              cor.coef = TRUE, cor.method = "pearson",
              xlab = "readz", ylab = "Nanost")
    
  }

   res[[length(res)+1]]<-df_g
}

print(res)

[[1]]
  sample_ID gene_ID    readz   Nanost       NEW
1        s1     TFT 6577.112 6582.497 100.08186
2        s2     TFT 6914.966 6192.676  89.55468
3        s3     TFT 7494.457 6508.501  86.84420
4        s4     TFT 7069.737 5966.418  84.39378

[[2]]
  sample_ID gene_ID    readz   Nanost       NEW
5        s1   Hbvrt 6346.545 7499.966 118.17399
6        s2   Hbvrt 7368.858 6860.801  93.10536
7        s3   Hbvrt 5671.581 5604.065  98.80957
8        s4   Hbvrt 6067.496 7420.354 122.29680 ## REMOVE THIS in the result and graph

[[3]]
   sample_ID gene_ID    readz   Nanost       NEW
9         s1    Myx4 5270.035 7086.622 134.47011
10        s2    Myx4 7338.199 5670.227  77.27002
11        s3    Myx4 5596.834 5595.212  99.97101
12        s4    Myx4 5477.589 7472.254 136.41502

[[4]]
   sample_ID gene_ID    readz   Nanost      NEW
13        s1  Rai56n 6526.715 6475.832 99.22040
14        s2  Rai56n 5512.179 5137.163 93.19660
15        s3  Rai56n 6109.446 5221.244 85.46182
16        s4  Rai56n 5836.242 5602.662 95.99776

我试过if else这样的声明：（在for循环中添加，但也没有用）

for ( g in unique(df$gene_ID)){
  
  df_g <- df[df$gene_ID == g, ]

### here to exclude a sample or more..
df_ex <- df[df$sample_ID != 's4' ,]

if (g == Hbvrt){ 
df_g <- df_ex
} else if (g != Hbvrt){
df_g <- df_g
}

.
.
.
.
. ## the same as before

提前非常感谢。

标签： r

解决方案

你可以试试 tidyverse

library(tidyverse)
df %>% 
  mutate(Nanost = runif(n(), 5000, 7500),
         NEW = Nanost / readz * 100) %>% 
  filter(!(gene_ID == "Hbvrt" &  sample_ID == "s4")) %>% 
  split(.$gene_ID)
$Hbvrt
sample_ID gene_ID    readz   Nanost       NEW
5        s1   Hbvrt 5745.199 6183.276 107.62510
6        s2   Hbvrt 5516.671 7123.757 129.13144
7        s3   Hbvrt 5521.244 5477.057  99.19969

$Myx4
sample_ID gene_ID    readz   Nanost       NEW
8         s1    Myx4 7379.206 5452.667  73.89233
9         s2    Myx4 6384.764 5388.520  84.39653
10        s3    Myx4 6974.163 6435.838  92.28116
11        s4    Myx4 5645.421 6328.313 112.09639

$Rai56n
sample_ID gene_ID    readz   Nanost       NEW
12        s1  Rai56n 6760.502 5206.982  77.02064
13        s2  Rai56n 5221.044 5927.277 113.52667
14        s3  Rai56n 5572.512 5535.466  99.33520
15        s4  Rai56n 6438.368 5470.014  84.95964

$TFT
sample_ID gene_ID    readz   Nanost       NEW
1        s1     TFT 7456.672 7049.609  94.54095
2        s2     TFT 6497.004 6943.117 106.86643
3        s3     TFT 5028.566 6971.517 138.63828
4        s4     TFT 6947.292 6704.789  96.50939

在基地R尝试

df_g$Nanost <- runif(nrow(df_g), 5000, 7500)
df_g$NEW <- df_g$Nanost / df_g$readz * 100 
gr = !(df_g$gene_ID == "Hbvrt" &  df_g$sample_ID == "s4")
df_list <- split(df_g[gr,], df_g$gene_ID[gr])

不需要循环。绘图可以使用lapply：

plots <- lapply(df_list, function(x) ggplot(x, aes(readz, NEW)) + geom_point())

但是如果你想坚持一个循环，那么只需在这里删除基因

for (g in unique(df$gene_ID[df_g$gene_ID != "Hbvrt"])){

推荐阅读