Looping a Function Over Multiple Data Files - Mindy Morales

Usefulness of skill

• DOM fluorescence spectroscopy generates many large matrix files
• Files require multiple corrections, calculations, and extraction of wavelength regions to provide useful information about DOM pool
• Usually done in MATLAB; open source code has not been developed for R
• Technique is not exclusive to EEM data and can be applied to any analysis that requires applying the same operation to multiple files.

Challenges

Working directory should only contain files of specified extension that you want to analyze

Sample code

In this example, a humification index (HIX, with corrections from Ohno 2008) is calculated from multiple .csv files

Define function

HIX<-function(low,high){sum(high)/(sum(high)+sum(low))}

Batch import csv files from wd

hixfiles <- list.files(pattern = "csv")

Create NULL vector to drop data into

hix_o <- NULL

Loop calculates HIX across all files in wd

for (i in 1:length(hixfiles)){

   tmp <-  read.csv(hixfiles[i])
   hix_o[i] = HIX(tmp$low, tmp$high)
}

Define vector to store results

results = data.frame(hixfiles, hix_o)

print(results)

Export to wd

write.csv(hix_o,file="hix.csv")