在使用ChemoSpec包前，用R计算重复样品数据的均值

Hey there! No need to apologize—we all start somewhere 😊. Absolutely, you can automate this workflow in R, which will save you tons of time as your sample count grows. Below is a straightforward, scalable solution tailored to your setup:

Step 1: Install & Load Required Packages #

We’ll use the tidyverse collection of packages (it makes reading files and manipulating data much more intuitive). If you haven’t installed it yet, run this first:

install.packages("tidyverse")
library(tidyverse)

Step 2: Read All CSV Files #

First, make sure all your 30 CSV files are in a single folder (you can set this folder as your working directory with setwd("path/to/your/folder")). Then we’ll read all files and add a column to track which original file each row came from:

# Get list of all CSV files in the working directory
file_list <- list.files(pattern = "\\.csv$")

# Read all files into a single data frame, adding a "file_id" column
all_data <- map_dfr(file_list, read_csv, .id = "file_id") %>%
  # Convert file_id from character (like "1") to numeric
  mutate(file_id = as.numeric(file_id))

Step 3: Group Replicates & Calculate Mean Absorbance #

Since your files are paired (1&2 = Sample 1, 3&4 = Sample 2, etc.), we’ll create a sample_id column to group each pair. Then we’ll calculate the mean absorbance for each wavelength within each sample:

mean_absorbance <- all_data %>%
  # Create sample_id: pair files into groups of 2
  mutate(sample_id = ceiling(file_id / 2)) %>%
  # Group by sample and wavelength, then calculate mean absorbance
  group_by(sample_id, wavelength) %>%
  summarize(mean_absorbance = mean(absorbance), .groups = "drop")

Step 4 (Optional): Export the Results to CSV #

If you want to save the averaged data for later use (or for ChemoSpec), you can export it to a CSV file:

write_csv(mean_absorbance, "averaged_ftir_data.csv")

How This Works with Your Example #

Using your simplified csv1 and csv2 data:

wavelength <- c(500, 550, 600)
absorbance <- c(2, 4, 3)
csv1 <- data.frame(wavelength, absorbance)
csv2 <- data.frame(wavelength, absorbance)

# Simulate the file reading step
all_data_example <- bind_rows(
  csv1 %>% mutate(file_id = 1),
  csv2 %>% mutate(file_id = 2)
)

# Calculate mean
mean_example <- all_data_example %>%
  mutate(sample_id = ceiling(file_id / 2)) %>%
  group_by(sample_id, wavelength) %>%
  summarize(mean_absorbance = mean(absorbance), .groups = "drop")

print(mean_example)

This will output exactly the mean values you calculated manually:

# A tibble: 3 × 3
  sample_id wavelength mean_absorbance
      <dbl>      <dbl>           <dbl>
1         1        500               2
2         1        550               4
3         1        600               3

For ChemoSpec Users #

If you plan to use this data directly in ChemoSpec, you can convert the mean_absorbance data frame into the format ChemoSpec expects. For example, pivot the data to have wavelengths as rows and samples as columns:

chemo_spec_ready <- mean_absorbance %>%
  pivot_wider(names_from = sample_id, values_from = mean_absorbance) %>%
  column_to_rownames("wavelength")

This approach scales seamlessly—even if you end up with 100 samples (200 CSV files), you won’t need to change any code!

内容的提问来源于stack exchange，提问作者G Jones