new file mode 100644

@@ -0,0 +1,44 @@

+

+

+# Must be 20nt long

+#spacers <- c("GGTGCTGATGCTGTGTGATG",

+#             "GGTGCTGATAAAGTGTGATG")

+.getModelScore <- function(spacers){

+    features <- .extractFeatures(spacers)

+    score <- model_offset + model_weights%*%t(features)

+    return(score)

+}

+

+model_weights <- c(0.14177385,

+                   0.06966514,

+                   0.04216254,

+                   0.03303432,

+                   0.02355430,

+                   -0.04746424,

+                   -0.04878001,

+                   -0.06981921,

+                   -0.07087756,

+                   -0.08160700)

+model_offset <- 0.6505037

+

+

+#' @importFrom Biostrings DNAStringSet letterFrequency

+.extractFeatures <- function(spacers){

+    gc <- rowSums(letterFrequency(DNAStringSet(substr(spacers,4,14)),

+                                  c("G", "C")))/10

+    spacers <- DNAStringSet(spacers)

+    mat <- as.matrix(DNAStringSet(spacers))

+    features <- list()

+    features[[1]] <- gc

+    features[[2]] <- mat[,20,drop=FALSE]=="G"

+    features[[3]] <- mat[,3,drop=FALSE]=="T" | mat[3]=="A"

+    features[[4]] <- mat[,12,drop=FALSE]=="G"| mat[12]=="A"

+    features[[5]] <- mat[,6,drop=FALSE]=="G"

+    features[[6]] <- mat[,4,drop=FALSE]=="T" | mat[4]=="A"

+    features[[7]] <- mat[,18,drop=FALSE]=="G" | mat[18]=="A"

+    features[[8]] <- mat[,5,drop=FALSE]=="C" | mat[5]=="A" 

+    features[[9]] <- mat[,14,drop=FALSE]=="G"

+    features[[10]] <- mat[,15,drop=FALSE]=="A"

+    features <- do.call(cbind,features)

+    return(features)

+}