Referencing a range of columns in dplyr

We can get the indexes first and then use rowSums,

v1 <- which(names(df) == 'A2') #find first column
#[1] 3
v2 <- which(names(df) == 'B1') #find last column
#[1] 4
df[rowSums(df[v1:v2])>10,]
#  txt A1 A2 B1 B2
#1 ala  6  9 12 23

I think the most dplyr-esque way would be:

df %>%
  filter(rowSums(select_(., 'A2:B1')) > 10)

Which gives:

#  txt A1 A2 B1 B2
#1 ala  6  9 12 23