CODEWITHSUNDEEP
rif-statementconditional-statements
sian
sian

Reputation: 77

R Filter column A based on whether specific value is in column B

I have a df where I want to filter for values in one column, based on if there is a specific value in the other column. For example I have a df like this:

x <- data.frame("Annotation" = c("lncRNA", "no lncRNA", "lncRNA", "no lncRNA", "lncRNA"), 
                "padj" = c(0.2, NA, 0.03, NA, 0.08))

A couple of things I've tried are

x <- x[ x$padj <= 0.1, ]

and

x <- x %>%
  filter(case_when(Annotation != "no lncRNA" ~ padj <= 0.1))

but neither give me the output I need. They produce these respectively

     Annotation padj
NA         <NA>   NA
3        lncRNA 0.03
NA.1       <NA>   NA
5        lncRNA 0.08 

  Annotation padj
1     lncRNA 0.03
2     lncRNA 0.08

instead of this which is what I'd like

  Annotation padj
1  no lncRNA   NA
2     lncRNA 0.03
3  no lncRNA   NA
4     lncRNA 0.08

Seems like there would be a simple solution, can anyone help? Thanks!

Upvotes: 0

Views: 271

Answers (2)

Paggles01
Paggles01

Reputation: 106

Could do something like this. May not be the most elegant solution.

sel <- x$padj <= 0.1
sel[is.na(sel)] <- TRUE
filtered <- x[sel, ]
row.names(filtered) <- seq_len(nrow(filtered))  # Only if you want to reset row names

Upvotes: 0

Shan R
Shan R

Reputation: 541

This will do it for you.

filter(x, padj <= 0.1 | is.na(padj))

Data:

x <- data.frame("Annotation" = c("lncRNA", "no lncRNA", "lncRNA", "no lncRNA", "lncRNA"), 
                "padj" = c(0.2, NA, 0.03, NA, 0.08), stringsAsFactors = F)

#Result:
#Annotation padj
#1  no lncRNA   NA
#2     lncRNA 0.03
#3  no lncRNA   NA
#4     lncRNA 0.08

Upvotes: 1

Related Questions