@@ -1,6 +1,6 @@

 Package: crisprDesign

 Title: Comprehensive design of CRISPR gRNAs for nucleases and base editors

-Version: 0.99.88

+Version: 0.99.89

 Authors@R: c(

     person("Jean-Philippe", "Fortin", email = "[email protected]", role = c("aut", "cre")),

     person("Luke", "Hoberecht", email = "[email protected]", role = c("aut"))

@@ -71,7 +71,7 @@ addOpsBarcodes <- function(guideSet,

 #'     diagonal distances be set to 0 to ignore self distances?

 #'     TRUE by default. 

 #' @param splitByChunks Should distances be calculated in a chunk-wise

-#'     manner? TRUE by default. Highly recommended when the set of query

+#'     manner? FALSE by default. Highly recommended when the set of query

 #'     barcodes is large to reduce memory footprint. 

 #' @param n_chunks Integer specifying the number of chunks to be used

 #'     when \code{splitByChunks=TRUE}. If NULL (default), number of chunks

@@ -118,7 +118,7 @@ getBarcodeDistanceMatrix <- function(queryBarcodes,

     if (is.null(min_dist_edit) & binnarize){

             stop("min_dist_edit must be specified when binnarize=TRUE.")

     }

-    if (!splitByChunks){

+    if (!splitByChunks | length(queryBarcodes<=200)){

         out <- .getChunkDistanceMatrix(queryBarcodes=queryBarcodes,

                                        targetBarcodes=targetBarcodes,

                                        min_dist_edit=min_dist_edit,

@@ -202,7 +202,10 @@ getBarcodeDistanceMatrix <- function(queryBarcodes,

 #'     have edit distances less than the min_dist_edit will not be

 #'     included in the library. 2 by default. 

 #' @param dist_method String specifying distance method. 

-#'     Must be either "hamming" (default) or "levenstein". 

+#'     Must be either "hamming" (default) or "levenstein".

+#' @param splitByChunks Should distances be calculated in a chunk-wise

+#'     manner? FALSE by default. Highly recommended when the set of query

+#'     barcodes is large to reduce memory footprint. 

 #' 

 #' @return A subset of the \code{df} containing the gRNAs

 #'     selected for the OPS library. 

@@ -229,7 +232,8 @@ designOpsLibrary <- function(df,

                              n_guides=4,

                              gene_field="gene",

                              min_dist_edit=2,

-                             dist_method=c("hamming","levenstein")

+                             dist_method=c("hamming","levenstein"),

+                             splitByChunks=FALSE

 ){

     dist_method <- match.arg(dist_method)

     df <- .validateOpsGrnaInput(df, gene_field)

@@ -243,12 +247,14 @@ designOpsLibrary <- function(df,

                      genes=genes)

     grnaList <- .initiateOpsLibrary(grnaList,

                                     dist_method=dist_method,

-                                    min_dist_edit=min_dist_edit)

+                                    min_dist_edit=min_dist_edit,

+                                    splitByChunks=splitByChunks)

     grnaList <- .updateOpsLibrary(grnaList,

                                   gene_field=gene_field,

                                   n_guides=n_guides,

                                   dist_method=dist_method,

-                                  min_dist_edit=min_dist_edit)

+                                  min_dist_edit=min_dist_edit,

+                                  splitByChunks=splitByChunks)

     out <- .getFinalOpsLibrary(grnaList)

     out <- out[order(out[[gene_field]], out[["rank"]]),,drop=FALSE]

     return(out)

@@ -273,6 +279,9 @@ designOpsLibrary <- function(df,

 #'     included in the library. 2 by default. 

 #' @param dist_method String specifying distance method. 

 #'     Must be either "hamming" (default) or "levenstein". 

+#' @param splitByChunks Should distances be calculated in a chunk-wise

+#'     manner? FALSE by default. Highly recommended when the set of query

+#'     barcodes is large to reduce memory footprint.

 #' 

 #' @author Jean-Philippe Fortin

 #'

@@ -308,7 +317,8 @@ updateOpsLibrary <- function(opsLibrary,

                              n_guides=4,

                              gene_field="gene",

                              min_dist_edit=2,

-                             dist_method=c("hamming","levenstein")

+                             dist_method=c("hamming","levenstein"),

+                             splitByChunks=FALSE

 ){

     dist_method <- match.arg(dist_method)

     df <- .validateOpsGrnaInput(df, gene_field)

@@ -321,7 +331,8 @@ updateOpsLibrary <- function(opsLibrary,

                                   gene_field=gene_field,

                                   n_guides=n_guides,

                                   dist_method=dist_method,

-                                  min_dist_edit=min_dist_edit)

+                                  min_dist_edit=min_dist_edit,

+                                  splitByChunks=splitByChunks)

     out <- .getFinalOpsLibrary(grnaList)

     out <- out[order(out[[gene_field]], out[["rank"]]),,drop=FALSE]

     return(out)

@@ -365,13 +376,15 @@ updateOpsLibrary <- function(opsLibrary,

 #' @importFrom Matrix rowSums

 .initiateOpsLibrary <- function(grnaList,

                                 dist_method,

-                                min_dist_edit

+                                min_dist_edit,

+                                splitByChunks

 ){

     selected <- grnaList[["selected"]]

     mat <- getBarcodeDistanceMatrix(queryBarcodes=selected[["opsBarcode"]],

                                     binnarize=TRUE,

                                     dist_method=dist_method,

-                                    min_dist_edit=min_dist_edit)

+                                    min_dist_edit=min_dist_edit,

+                                    splitByChunks=splitByChunks)

     good <- Matrix::rowSums(mat>0)==0

     # In case all guides are "bad", add first one only:

     if (sum(good)==0){

@@ -390,7 +403,8 @@ updateOpsLibrary <- function(opsLibrary,

                               gene_field,

                               n_guides,

                               dist_method,

-                              min_dist_edit

+                              min_dist_edit,

+                              splitByChunks

 ){

     shouldWeContinue <- TRUE

     while (shouldWeContinue){

@@ -399,7 +413,8 @@ updateOpsLibrary <- function(opsLibrary,

                                           gene_field=gene_field,

                                           n_guides=n_guides,

                                           dist_method=dist_method,

-                                          min_dist_edit=min_dist_edit)

+                                          min_dist_edit=min_dist_edit,

+                                          splitByChunks=splitByChunks)

         counts <- table(factor(grnaList[["selected"]][[gene_field]],

                           levels=grnaList[["genes"]]))

         incomplete <- names(which(counts<n_guides))

@@ -425,7 +440,8 @@ updateOpsLibrary <- function(opsLibrary,

                                   gene_field,

                                   n_guides,

                                   dist_method,

-                                  min_dist_edit

+                                  min_dist_edit,

+                                  splitByChunks

 ){

     .getCandidates <- function(genes, n){

@@ -451,7 +467,8 @@ updateOpsLibrary <- function(opsLibrary,

         # most divergent:

         dist <- getBarcodeDistanceMatrix(cands[["opsBarcode"]],

                                          dist_method=dist_method,

-                                         min_dist_edit=min_dist_edit)

+                                         min_dist_edit=min_dist_edit,

+                                         splitByChunks=splitByChunks)

         score <- Matrix::rowSums(dist>0)

         cands <- cands[order(score),,drop=FALSE]

@@ -494,7 +511,8 @@ updateOpsLibrary <- function(opsLibrary,

         dist <- getBarcodeDistanceMatrix(cands[["opsBarcode"]],

                                          lib[["opsBarcode"]],

                                          dist_method=dist_method,

-                                         min_dist_edit=min_dist_edit)

+                                         min_dist_edit=min_dist_edit,

+                                         splitByChunks=splitByChunks)

         cands <- cands[Matrix::rowSums(dist)==0,,drop=FALSE]

         grnaList <- .incrementalUpdate(grnaList, cands)

     }

@@ -9,7 +9,8 @@ designOpsLibrary(

   n_guides = 4,

   gene_field = "gene",

   min_dist_edit = 2,

-  dist_method = c("hamming", "levenstein")

+  dist_method = c("hamming", "levenstein"),

+  splitByChunks = FALSE

 )

 }

 \arguments{

@@ -29,6 +30,10 @@ included in the library. 2 by default.}

 \item{dist_method}{String specifying distance method. 

 Must be either "hamming" (default) or "levenstein".}

+

+\item{splitByChunks}{Should distances be calculated in a chunk-wise

+manner? FALSE by default. Highly recommended when the set of query

+barcodes is large to reduce memory footprint.}

 }

 \value{

 A subset of the \code{df} containing the gRNAs

@@ -37,7 +37,7 @@ diagonal distances be set to 0 to ignore self distances?

 TRUE by default.}

 \item{splitByChunks}{Should distances be calculated in a chunk-wise

-manner? TRUE by default. Highly recommended when the set of query

+manner? FALSE by default. Highly recommended when the set of query

 barcodes is large to reduce memory footprint.}

 \item{n_chunks}{Integer specifying the number of chunks to be used

@@ -10,7 +10,8 @@ updateOpsLibrary(

   n_guides = 4,

   gene_field = "gene",

   min_dist_edit = 2,

-  dist_method = c("hamming", "levenstein")

+  dist_method = c("hamming", "levenstein"),

+  splitByChunks = FALSE

 )

 }

 \arguments{

@@ -32,6 +33,10 @@ included in the library. 2 by default.}

 \item{dist_method}{String specifying distance method. 

 Must be either "hamming" (default) or "levenstein".}

+

+\item{splitByChunks}{Should distances be calculated in a chunk-wise

+manner? FALSE by default. Highly recommended when the set of query

+barcodes is large to reduce memory footprint.}

 }

 \value{

 A data.frame containing the original gRNAs from