Assigning column names to df, stop characters from automatically converting to numbers?

Question

I have a df of gene names and a df of values for those genes. I want to combine the two (label the data frame with the gene names).

I have a 1x23000 df, "genes":

AB1G, A1CF, ..., ZYY1

and a 8000x23000 df, "expression":

1 0 ... 3

0 0 ... 1

2 2 ... 0

I want to combine them into one df:

A1BG A1CF ... ZYY1

1 0 ... 3

0 0 ... 1

2 2 ... 0

When I use rbind, the gene names are maintained but all non-zero values in "expression" turn to N/A.

When I use colnames(expression)=genes or colnames(expression)=paste(genes), the expression numbers are maintained but the gene names are converted to seemingly meaningless numbers.

Upon further investigation, I saw that the structure of "genes" for some reason is "factor" - when I convert with as.character, all genes get reassigned what seems like a random number (AB1G is now 1143, A1CF is now 967, etc).

I think there may be an issue with incompatible formats of the gene and expression dfs. How can I add character column names to a numeric df?

May be you need colnames(expression) <- as.character(unlist(genes)) or colnames(expression) <- c(as.matrix(genes)) — akrun
– akrun, Commented Oct 23, 2019 at 17:48
try to create a reproducible example with basic R code giving us a couple of dataframes that maybe is a 5 row sample of your data. We need something to be able to reproduce on our side in order to help or understand what your issue is. also create an example of your ideal state post fix. — Matt W.
– Matt W., Commented Oct 25, 2019 at 15:23

akrun · Accepted Answer · 2019-10-23 17:54:05Z

0

The issue of appearance of numeric value instead of string is because the columns were factors and when it got coerced to integer storage mode values. Instead, we can convert to character class after unlisting or convert to matrix which automatically convert the class to character and then remove the dim attributes with c (to convert to vector)

 colnames(expression) <- as.character(unlist(genes))

Or

colnames(expression) <- c(as.matrix(genes))

answered Oct 23, 2019 at 17:54

akrun

891k38 gold badges590 silver badges700 bronze badges

Sign up to request clarification or add additional context in comments.

Collectives™ on Stack Overflow

Assigning column names to df, stop characters from automatically converting to numbers?

1 Answer 1

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

Comments

Your Answer

Sign up or log in

Post as a guest

Related