how to remove duplicates from a data frame in R

Question

I have a data frame of correlation coefficients like the following. In the data frame it has correlation coefficients of a*b and b*a which are the same. How do I remove this duplicates? Can anyone please help

**Var1, Var2, r**
ApoA1.ng.ml.1, Apo.B.ng.ml, 0.9998438
Apo.B.ng.ml, ApoA1.ng.ml.1, 0.9998438
SLM.T0., TBW.T0., 0.9992563
TBW.T0., SLM.T0., 0.9992563
Insulin.mercdiaConc..U.L, Insulin..pg.ml, 0.9313702
Insulin..pg.ml, Insulin.mercdiaConc..U.L, 0.9313702

Answer 1

If the other techniques don't quite work, you can use temporary min/max strings and de-duplicated from those:

x <- read.csv(stringsAsFactors=FALSE, text="
Var1,Var2,r
ApoA1.ng.ml.1,Apo.B.ng.ml,0.9998438
Apo.B.ng.ml,ApoA1.ng.ml.1,0.9998438
SLM.T0.,TBW.T0.,0.9992563
TBW.T0.,SLM.T0.,0.9992563
Insulin.mercdiaConc..U.L,Insulin..pg.ml,0.9313702
Insulin..pg.ml,Insulin.mercdiaConc..U.L,0.9313702")

x[!duplicated(pmin(x$Var1, x$Var2),pmax(x$Var1, x$Var2)),]
#                       Var1           Var2         r
# 1            ApoA1.ng.ml.1    Apo.B.ng.ml 0.9998438
# 3                  SLM.T0.        TBW.T0. 0.9992563
# 5 Insulin.mercdiaConc..U.L Insulin..pg.ml 0.9313702

(You can also assign them temporarily to columns in the frame, ala

x$m1 <- pmin(x$Var1, x$Var2)
x$m2 <- pmax(x$Var1, x$Var2)
x[!duplicated(x[c("m1","m2")]),]

though you then have to remove the temp variables yourself.)

Answer 2

We could try using the sqldf package here:

library(sqldf)
sql <- "SELECT MIN(Var1, Var2), MAX(Var2, Var1), MAX(r) AS R
        FROM df
        GROUP BY MIN(Var1, Var2), MAX(Var2, Var1)"

df_out <- sqldf(sql)

Demo