How to add column with matching gene symbols for gene probe ID in R

Question

[myyid68]

I have a dataframe with genes and samples (cancer vs. normal) and I already did LASSO and cross validation to choose best lambda, as well as finding the genes with non-zero coefficients (x in code below is my dataframe containing these). What I want to do next is add another column to x which contains the gene symbols (symbol from original dataframe daf) corresponding to those gene with non-zero coefficients in x such that my final dataframe with have probe_id as first column, followed by symbol and finally, the last column contains the non-zero coefficients. I realise this is probably super basic but I have been trying for over an hour to figure out a way to make this work but haven't had success. Any suggestions on what is the best way to do that? Below is my code:

probeID<-c("213456_at", "217428_s_at", "219230_at", "226228_at","230030_at")
symbol<-c("SOSTDC1","COL10A1", "TMEM100", "AQP4", "HS6ST2")
BCR1<-c(28.005966, 30.806433, 17.341375, 17.40666, 30.039436)
BCR2<-c(30.973469, 29.236025, 30.41161, 20.914383, 20.904331)
BCR3<-c(26.322796, 25.542833, 22.460772, 19.972183, 30.409641)
BCR4<-c(26.441898, 25.837685, 23.158352, 20.379173, 33.81327)
BCR5<-c(39.750206, 19.901133, 28.180124, 22.668673, 25.748884)
CTL6<-c(23.004385, 28.472675, 23.81621, 26.433413, 28.851719)
CTL7<-c(22.239546, 28.741674, 23.754929, 26.015385, 28.16368)
CTL8<-c(29.590443, 30.041988, 21.323061, 24.272501, 18.099016)
CTL9<-c(15.856442, 22.64224, 29.629637, 25.374926, 22.356894)
CTL10<-c(38.137985, 24.753338, 26.986668, 24.578161, 19.223558)
daf<-data.frame(probeID, symbol, BCR1, BCR2, BCR3, BCR4, BCR5, CTL6, CTL7, CTL8, CTL9, CTL10)
daf1<-t(daf[,3:12])
colnames(daf1)<-daf$probeID
Type<-c("cancer", "cancer", "cancer", "cancer", "cancer", "normal", "normal", "normal", "normal", "normal")
Sample<-c("BCR1", "BCR2", "BCR3", "BCR4", "BCR5", "CTL6", "CTL7", "CTL8", "CTL9", "CTL10")
type.df<-data.frame(Sample, Type)
daf2<-data.frame(daf1, type.df$Type)
names(daf2)[names(daf2) == "type.df.Type"] <- "Type"
daf2$Type<-as.factor(daf2$Type)
lassoModel <- glmnet(
 x=data.matrix(daf2[,-6]),
 y=daf2$Type,
  alpha=1,
  family="binomial")
cv.lassoModel<- cv.glmnet(
  x=data.matrix(daf2[,-6]),
  y=daf2$Type,
 alpha=1, family="binomial")

idealLambda <- cv.lassoModel$lambda.min
co <- coef(cv.lassoModel, s=idealLambda, exact=TRUE)
cv.glm.probe<-coef(cv.lassoModel, s="lambda.min")
x<-data.frame(cv.glm.probe[cv.glm.probe[,1]!=0,])

Answer 1

[dsull]

First: in the line daf2<-data.frame(daf1, type.df$Type), you a creating a new dataframe, called daf2, which will have column names that start with numbers (because daf has column names, e.g. 213456_at, that start with numbers). Creating a new data frame with column names that start with numbers is a little bit problematic for R (the letter 'X' will get prepended to those column names), so you need to add the parameter: check.names=FALSE (see below):

daf2<-data.frame(daf1, type.df$Type, check.names=FALSE)
names(daf2)[names(daf2) == "type.df$Type"] <- "Type"

Second: your final dataframe x has row names that include the Intercept coefficient as well as the coefficients for your non-zero probes. There is only one column which consists of the coefficients. So let's do the following to have two columns: your probe ID columns and your coefficients column:

x <- data.frame(rownames(x), x)
colnames(x) <- c("probeID", "coefficient")

Third: Alright, are you with me? Now, it's time to add our final column (the symbol column). We do this by a left outer join (which basically preserves everything your x data frame while combining it with another data frame containing your gene symbols; note that we're merging the two data frames based on their mutual probe IDs). And then we arrange the final data frame so we have the columns in the following order (as you desire): probeID, symbol, coefficient

x <- merge(x, daf[,c("probeID", "symbol")], by="probeID", all.x=TRUE)
x <- x[,c("probeID", "symbol", "coefficient")]

Finally: After putting it all together, when you print out x, you should get something like the following output:

      probeID  symbol  coefficient
1 (Intercept)    <NA> -41.23471919
2 217428_s_at COL10A1   0.18134947
3   226228_at    AQP4   1.61933359
4   230030_at  HS6ST2  -0.03797544