CODEWITHSUNDEEP
rdataframe
Evan
Evan

Reputation: 1499

How to delete a row in R that doesn't have a number

My data looks like:

SNP             A1      A2      EFF                     FRQ
rs12565286      C       G       -0.00225985777786465    .04354
rs11804171      A       T       -0.00530020318295282    .04485
rs3094315       A       G       -0.0042551489236695     .8364
rs12562034      A       G       -0.00911972489527125    .09763
rs12124819      A       G       0.0250148724382224      .
rs2980319       A       T       0.0178927256033542      .1306
rs4040617       A       G       -0.0173263263037023     rabbit

I would like to delete the rows that contain "." and that contain "rabbit". Is there a way to keep the numbers? The reason I want to do this is because I'm using a large data file for a manipulation and I am getting a 1 - gwas.data$FRQ[i] : non-numeric argument to binary operator error. Am I right in assuming the error is because some data isn't a number? I haven't even checked to see if the data has non-numeric values in the column because it's 3 million lines long.

Upvotes: 1

Views: 1219

Answers (1)

DunderChief
DunderChief

Reputation: 726

Example data.frame: 

df <- data.frame(a=1:10, b=1:10, FRQ=c(rnorm(8), '.', 'rabbit'), stringsAsFactors=FALSE)

To check the class of all your columns try: lapply(df, class)

If the FRQ column is character, you can convert it to numeric by removing all non-numerics, then convert to numeric. Like this:

library(stringr)
df <- df[!str_detect(df$FRQ, '([A-Za-z])'), ]
df <- df[!str_detect(df$FRQ, '\\.$'), ]
df$FRQ <- as.numeric(df$FRQ)

Upvotes: 2

Related Questions