diff --git a/Approach2.R b/Approach2.R
new file mode 100644
index 0000000..1139253
--- /dev/null
+++ b/Approach2.R
@@ -0,0 +1,62 @@
+
+setwd("/Users/sayyar/Downloads")
+setwd("/Users/sayyar/Documents/openpredict")
+library("splitstackshape")
+
+###Reading the data
+data.chem.1.3 <- data.frame(read.delim(file="data/features/drugbank-sideeffects-similarity.tab", sep="\t",header=F))
+tbDrugs <- c("DB00679","DB00845","DB00479", "DB00601", "DB01044", "DB01137", "DB00218", "DB01165", "DB01044", "DB00766", "DB00615", "DB04934", "DB01045", "DB01201", "DB01082", "DB00609", "DB00951", "DB00609", "DB00760", "DB00679", "DB00314", "DB01172", "DB00339", "DB00314", "DB00260",  "DB00760","DB00330", "DB00494")
+data.chem.1.6 <- data.frame(read.delim(file="data/features/drugbank-sider-smiles-similarity-MACCS.tab",sep="\t",header=F))
+new.Drugs <- data.frame(read.delim(file = "DiseaseDrugs.csv", sep= "\t", header = F))
+
+###Functions for data cleanup
+rowname.data.chem <- function(data)
+{
+  rowname       <- apply(data[,1:2],1,function(x) paste(sort(c(x[1],x[2]))[1], sort(c(x[1],x[2]))[2] ,sep="_"))
+  mat           <- as.matrix(data[,3])
+  rownames(mat) <- rowname
+  return(mat)
+}
+
+splitAndOrder <-function(x) {
+  x=unlist(strsplit(x, "_"))
+  x=paste(sort(x), collapse = "_")
+  return(x)
+}
+
+
+###Filter the Sider similarity data through known drugs
+drugs.tb.known <- sort(unique(tbDrugs))
+
+drugs.1 <- data.chem.1.3[data.chem.1.3$V1 %in% tbDrugs,]
+drugs.2 <- data.chem.1.3[data.chem.1.3$V2 %in% tbDrugs,]
+
+drugs.tb <- rbind(drugs.1,drugs.2)
+drugs.tb <- unique(drugs.tb)
+
+drugs.tb <- drugs.tb[which(drugs.tb$V3 < 0.3), ]
+d <- cSplit(new.Drugs, "V1", ":")
+head(d)
+new.Drugs <- d$V1_2
+new.Drugs <- gsub(']','', new.Drugs)
+
+new.Drugs - new.Drugs[!is.na(new.Drugs)]
+
+######Filter the drug drug side effect similarities from TB with candidate drugs
+candidate.Drugs <- new.Drugs
+
+sim.drugs.1 <- as.character(drugs.tb$V1[drugs.tb$V1 %in% candidate.Drugs])
+sim.drugs.2 <- as.character(drugs.tb$V2[drugs.tb$V2 %in% candidate.Drugs])
+
+sim.drugs <- c(sim.drugs.1, sim.drugs.2)
+sim.drugs <- sort(unique(sim.drugs))
+
+known.unknown.tb <- apply(expand.grid(sim.drugs, drugs.tb.known),1,function(x) paste(x[1],x[2],sep="_"))
+known.unknown.tb <- sapply(known.unknown.tb, splitAndOrder, USE.NAMES = F)
+
+
+####Filter the chemical similarity MACCS data with the above candidate drug list
+results.1.data <- data.chem.1.6[rownames(data.chem.1.6) %in% known.unknown.tb, ]
+results.1.data <- data.frame(results.1.data)
+
+results.1.data <- results.1.data[order(-results.1.data$score), , drop = FALSE]