What to do when your column name is "NA"...

[matt.curcio.ri]

I have a data file with protein amino acids info. (actually dipeptides) as features. The dipeptide in question is "NA". To me "NA" = Asparagine + Alanine. However when R sees "NA" it calls the feature X244.

Running: colnames(cancer_human_comps)[244] <- "NA" # This does not clear the issue.

Any ideas?

[andresrcs]

@matt.curcio.ri

However when R sees "NA" it calls the feature X244.

When does this happen? Could you ask this with a minimal REPRoducible EXample (reprex)?

[Matthias]

How do you get the data into R? It works when starting from an Excel sheet using read_excel from the readxl library.
This puts the NA column in backticks.

Maybe other importers can be modified as well.
For example, read_csv(readr / tidyverse) accepts definitions for NA, when you say:
dipeptides = read_csv("Book2.csv", na = "")
you basically excludes "NA" as NA and then its kept as well.

When you have NA in your matrix then you have another problem.

dipeptides = read_csv("Book2.csv", na = "")
Parsed with column specification:
cols(
  Sample = col_character(),
  AB = col_double(),
  BC = col_double(),
  GH = col_double(),
  JI = col_double(),
  KL = col_double(),
  `NA` = col_double(),
  DG = col_double()
)
>

[matt.curcio.ri]

Working on it... Reprex render is hanging...

[matt.curcio.ri]

Try this...

However this seems to work. Issue solved.