对于当前目录中的所有文件，请在另一个目录中查找具有相同前缀的文件。 [R

Question

I have a very basic question, and I apologize if it is asked elsewhere (I tried to find the answer, I really did).

I have written a script that creates two directories, and saves information for a given file in each directory under the same name. In one directory I have formated data to make a boxplot using ggplot, in the other directory I have saved annotation information. I then make a boxplot, and I want to search the annotation directory for the corresponding annotation file so I can add the annotations to the boxplot. The code is setup to perform this operation for "all files" in a given directory, so I cant simply change working directories and load the annotation file by name. Here is what I got:

In the directory called ggplot/data, files are saved as: my_data_1.csv

In the directory called ggplot/annotation, files are saved as: my_data_1.csv

Final annotated graphs are saved in ggplot/graph_output.

# goto ggplot data directory
setwd("/home/path/to/ggplot/data")

#look for all files
inFilePaths = list.files(path=".", pattern=glob2rx("*"), full.names=TRUE)

#make a ggplot2 boxplot for every file with
for (inFilePath in inFilePaths)
{ 
  # Read in each data file as a dataframe
  inFileData = read_csv(inFilePath)

  # Make a ggplot. **This is only part of my code to save space**
  plot1 = ggplot(data =inFileData, mapping= aes(x=Sample, y=Expression)) +
    scale_fill_manual(values=c("#606060", "#29a329"))

  # Change directories to annottaion folder
  setwd("/home/path/to/ggplot/annotation")

  ####Help!!!!#### Write something to find the file with same inFilePath name to get annotations
  ##Maybe something like this:
  inFilePaths2 = list.files(path=".", pattern=glob2rx(inFileData), full.names=TRUE)
  ##This does not work because it cant find the same inFileData file used to make the ggplot

  # annotate gglot with corresponding annotation file
  for (inFilePath in FilePaths2)
  {
    palues = read_csv(...of the file that matches the file name of the ggplot data) 
    plt2_annot <- plot1 +
      geom_text(data=pvalues, aes(x=value, y=breaks,label = paste('P:',format.pval(pval, digits=1))))
  }


  # specify size of ggplot base on number of boxes displayed using total rows of data
  n = 0.25+(0.75*(nrow(unique(select(inFileData, Gene)))))

  # Change directories to graph output folder, and save graph
  setwd("/home/path/to/ggplot/graph_output")
  ggsave(filename = paste(inFilePath, ".png"), plot=plot2, height = 1.5, width = n, units = "in")
}

Answer 1

Using Gregor's comments, I managed to come up with a very easy solution.

1) I changed how I named my files in each directory so the data file and corresponding annotation file have the exact same name.

2) Rather than trying to implement some function to find the corresponding annotation file to the current inFilePath data file, simply changing the directory to the annotation directory and re-loading the inFilePath using read_csv(inFilePath) resulted in loaded the corresponding annotation file.

Here is the code that ended up working for me:

# goto ggplot data directory
setwd("/home/path/to/ggplot/data")

#look for all files
inFilePaths = list.files(path=".", pattern=glob2rx("*"), full.names=TRUE)

#make a ggplot2 boxplot for every data file
for (inFilePath in inFilePaths)
{ 
  #Need to set directory again due to directory change lower in the loop
  setwd("/home/path/to/ggplot/data")
  # Read in each data file as a dataframe
  inFileData = read_csv(inFilePath)

  #check to see which data is loaded
  print(inFileData)

  #make a ggplot from the ggplot data

  # Change directories to annotation folder
  setwd("/home/path/to/ggplot/annotation")

  #load new annotation data. The file names are the same, so loading the same file name in the annotations
  # directory actually loads the annotations for the corresponding plot
  inFileData2 = read_csv(inFilePath)

  #check to make sure the correct annotation file is loaded
  print(inFileData2)

  #add annotation to ggplot graph
  #now that I can access the correct annotation, I'll work on this part next.
  #then save the graph

  }

Thanks for the help.