I made the code for chisquare-test as below in R

# Load the necessary libraries
library(readxl)
library(tidyverse)

# Read the data from the Excel file
data <- read_excel('Chisquare_test.xlsx', sheet = "Sheet1")

# Extract the taxa names
taxa_names <- colnames(data)[-c(1, 20)]

# Create an empty matrix to store p-values
num_taxa <- length(taxa_names)
p_values <- matrix(NA, nrow = num_taxa, ncol = 1)

# Loop through the taxa and conduct a chi-square test for each taxon
for (i in 1:num_taxa) {
  # Create a contingency table for the taxon and the categorical variable
  contingency_table <- table(data[, i + 2], data[, 20])

  # Conduct the chi-square test
  chi_sq_result <- chisq.test(contingency_table)$p.value

  # Store the p-value in the matrix
  p_values[i, 1] <- chi_sq_result
}

# Create row names for the matrix
rownames(p_values) <- taxa_names

# Print the p-values matrix
print(p_values)

My raw data in excel 1st column is ID, 20th column is categorical variable. I want 18 chi-square results from taxon using for loop easily.

However, the error

Error in xtfrm.data.frame(x) : 
  (converted from warning) cannot xtfrm data frames

after the line contingency_table <- table(data[, i + 2], data[, 20])

I found that both data[, i + 2] and data[, 20] have the same length in the raw file, but R says they're different.

1. How should I think this (I can't understand why the error happens) and revise my code?
2. I found that chisq.test(data$Acanthamoeba,data$Water) works but chisq.test(data[,2],data[,20]) doesn't work - specifying first columns works, but indicating data column sequence doesn't work. Why is that?

The raw material can be found from the link below

https://docs.google.com/spreadsheets/d/1EPah3-aoGf7hvqpbS-w_UiguSKEQihpE/edit?usp=sharing&ouid=108716001644013408238&rtpof=true&sd=true