@@ -63,7 +63,9 @@ Imports:

     methods,

     S4Vectors,

     dplyr,

-    stats

+    stats,

+    glue,

+    BiocGenerics

 Depends:

     R(<= 3.4.2)

 VignetteBuilder: knitr

@@ -92,6 +94,7 @@ Collate:

     'gmql_cover.R'

     'gmql_difference.R'

     'gmql_extend.R'

+    'gmql_group.R'

     'gmql_join.R'

     'gmql_map.R'

     'gmql_materialize.R'

@@ -67,6 +67,7 @@ exportMethods(collect)

 exportMethods(cover)

 exportMethods(extend)

 exportMethods(filter)

+exportMethods(group_by)

 exportMethods(map)

 exportMethods(merge)

 exportMethods(select)

@@ -78,10 +79,16 @@ import(httr)

 import(xml2)

 importClassesFrom(GenomicRanges,GRangesList)

 importClassesFrom(S4Vectors,DataTable)

+importFrom(BiocGenerics,setdiff)

+importFrom(BiocGenerics,union)

 importFrom(GenomicRanges,makeGRangesFromDataFrame)

+importFrom(S4Vectors,aggregate)

+importFrom(S4Vectors,merge)

 importFrom(S4Vectors,metadata)

 importFrom(data.table,fread)

 importFrom(dplyr,bind_cols)

+importFrom(dplyr,filter)

+importFrom(glue,glue)

 importFrom(methods,is)

 importFrom(methods,isClass)

 importFrom(methods,new)

@@ -1,15 +1,3 @@

-#' Method aggregate

-#' 

-#' Wrapper to GMQL merge function

-#' 

-#' @name aggregate

-#' @rdname aggregate-GMQLDataset-method

-#' @aliases aggregate

-#'  

-setGeneric("aggregate", function(x, ...) 

-                                    standardGeneric("aggregate"))

-

-

 #' Method filter

 #' 

 #' Wrapper to GMQL select function

@@ -71,7 +59,6 @@ setGeneric("take", function(data, ...) standardGeneric("take"))

 #' @aliases extend GMQLDataset-method

 setGeneric("extend", function(.data, ...) standardGeneric("extend"))

-

 #' Method select

 #' 

 #' Wrapper to GMQL project function

@@ -81,6 +68,7 @@ setGeneric("extend", function(.data, ...) standardGeneric("extend"))

 #' @aliases select 

 setGeneric("select", function(.data, ...) standardGeneric("select"))

+

 #' Method arrange

 #' 

 #' Wrapper to GMQL order function

@@ -94,3 +82,12 @@ setGeneric("arrange", function(.data, metadata_ordering = NULL,

                     reg_fetch_opt = NULL, reg_num_fetch = 0, ...) 

                             standardGeneric("arrange"))

+#' Method group_by

+#' 

+#' Wrapper to GMQL group function

+#' 

+#' @name group_by

+#' @rdname group_by-GMQLDataset-method

+#' @aliases group_by 

+setGeneric("group_by", function(.data, ...) standardGeneric("group_by"))

+

@@ -1,9 +1,9 @@

-#' FILTER AND EXTRACT 

+#' FILTER AND EXTRACT

 #'

-#' This function let user to create a new GRangeslist with fixed information: 

-#' seqnames,ranges ans strand and a variable part made up by the regions 

+#' This function let user to create a new GRangeslist with fixed information:

+#' seqnames,ranges ans strand and a variable part made up by the regions

 #' defined as input.

-#' The metadata and metadata_prefix are used to filter the data and choose 

+#' The metadata and metadata_prefix are used to filter the data and choose

 #' only the samples that match at least one metdatata with its prefix.

 #' The input regions are shown for each sample obtained from filtering.

 #'

@@ -12,51 +12,57 @@

 #' @importFrom data.table fread

 #' @importFrom rtracklayer import

 #'

-#' @param data string GMQL dataset folder path or GrangesList 

+#' @param data string GMQL dataset folder path or GrangesList

 #' object

-#' @param metadata vector of character containing names of metadata 

+#' @param metadata vector of character containing names of metadata

 #' to be searched for in metadata files.

 #' data will be extracted if at least one condition is satisfied:

 #' this condition will be logically "ANDed" with prefix filtering (see below)

-#' if NULL no filtering action occured 

+#' if NULL no filtering action occured

 #' (i.e every sample will be taken for regions filtering)

-#' @param metadata_prefix vector of character that will filter metadata 

+#' @param metadata_prefix vector of character that will filter metadata

 #' containing rispectively every element of this vector.

 #' number of elelment in both vector must match

-#' @param regions vector of character that will extract only region 

-#' attribute specified; if NULL no regions will be taken and the output 

+#' @param regions vector of character that will extract only region

+#' attribute specified; if NULL no regions will be taken and the output

 #' will be only GRanges made up by the first attribute

 #' (seqnames,start,end,strand)

+#' @param suffix name for each metadata column of GRanges. by default is the 

+#' "antibody_target". This string is taken from sample metadata file or from

+#' metadata() associated. If not present, the column name is the name of 

+#' selected regions

 #'

-#' 

-#' @details 

-#' This function works only with datatset or Grangeslist that have the same 

+#' @details

+#' This function works only with datatset or Grangeslist that have the same

 #' information about regions attribute (but of course different value)

-#' 

+#' In case of Grangeslist data input the function will search for metadata

+#' into metadata() function associated to Grangeslist.

+#'

 #' @return granges with selected regions (if any) in elementMetadata

 #'

 #' @examples

 #'

 #' test_path <- system.file("example", "DATASET", package = "RGMQL")

 #' filter_and_extract(test_path,regions = c("pvalue", "peak"))

-#' 

+#'

 #' grl = import_gmql(test_path, TRUE)

 #' filter_and_extract(grl, regions = c("pvalue", "peak"))

 #'

 #'

 #' @export

 #'

-filter_and_extract <- function(data, metadata = NULL, 

-                                metadata_prefix = NULL, regions = NULL)

+filter_and_extract <- function(data, metadata = NULL,

+                                metadata_prefix = NULL, regions = NULL, 

+                                suffix = "antibody_target")

 {

     if(is(data,"GRangesList"))

-        .extract_from_GRangesList(data,metadata,metadata_prefix,regions)

+        .extract_from_GRangesList(data, metadata, metadata_prefix, regions)

     else

-        .extract_from_dataset(data,metadata,metadata_prefix,regions)

+        .extract_from_dataset(data, metadata, metadata_prefix, regions, suffix)

 }

-.extract_from_dataset <- function(datasetName, metadata = NULL, 

-                        metadata_prefix = NULL, regions = NULL)

+.extract_from_dataset <- function(datasetName, metadata, metadata_prefix, 

+                                        regions, suffix)

 {

     datasetName <- sub("/*[/]$","",datasetName)

     if(basename(datasetName) !="files")

@@ -64,55 +70,66 @@ filter_and_extract <- function(data, metadata = NULL,

     if(!dir.exists(datasetName))

         stop("Directory does not exists")

-

+    

     gdm_meta_files <- list.files(datasetName, pattern = "*.gdm.meta$",

                                     full.names = TRUE)

     gtf_meta_files <- list.files(datasetName, pattern = "*.gtf.meta$",

                                     full.names = TRUE)

-

+    

     if(length(gdm_meta_files)==0 && length(gtf_meta_files)==0)

         stop("no samples present or no files format supported")

-

+    

     if(length(gdm_meta_files)>=1 && length(gtf_meta_files)>=1)

         stop("GMQL dataset cannot be mixed dataset: no GTF and GDM together")

-

+    

     vector_field <- .schema_header(datasetName)

-

+    

     if(length(gdm_meta_files)>0)

     {

-        samples_file <- .check_metadata_files(metadata,metadata_prefix,

-                                                gdm_meta_files)

+        samples_with_suffix <- .check_metadata_files(metadata,metadata_prefix,

+                                                gdm_meta_files, suffix)

+        

+        samples_file <- sapply(samples_with_suffix, function(x) x$sample)

+        suffix_vec <- sapply(samples_with_suffix, function(x) x$suffix)

+        suffixes <- unlist(suffix_vec)

         samples_to_read <- unlist(samples_file)

+        

         if(length(samples_to_read)>0)

             samples_to_read <- gsub(".meta$", "", samples_to_read)

         else

             samples_to_read <- gsub(".meta$", "", gdm_meta_files)

-

-        granges <- .parse_gdm_files(vector_field,samples_to_read,regions)

+        

+        granges <- .parse_gdm_files(vector_field,samples_to_read,regions,

+                                                        suffixes)

     }

     else

     {

-        samples_file <- .check_metadata_files(metadata,metadata_prefix,

-                                                gtf_meta_files)

+        samples_with_suffix <- .check_metadata_files(metadata,metadata_prefix,

+                                                    gtf_meta_files, suffix)

+        

+        samples_file <- sapply(samples_with_suffix, function(x) x$sample)

+        suffix_vec <- sapply(samples_with_suffix, function(x) x$suffix)

+        suffixes <- unlist(suffix_vec)

         samples_to_read <- unlist(samples_file)

+        

         if(length(samples_to_read)>0)

             samples_to_read <- gsub(".meta$", "", samples_to_read)

         else

             samples_to_read <- gsub(".meta$", "", gtf_meta_files)

-

-        granges <- .parse_gtf_files(samples_to_read,regions)

+        

+        granges <- .parse_gtf_files(samples_to_read, regions, suffixes)

     }

 }

-.extract_from_GRangesList <- function(rangesList, metadata = NULL,

-                                metadata_prefix = NULL, regions = NULL)

+.extract_from_GRangesList <- function(rangesList, metadata, metadata_prefix, 

+                                        regions)

 {

     if(!is(rangesList,"GRangesList"))

         stop("only GrangesList admitted")

-

+    

     if(length(rangesList)<=0)

         stop("rangesList empty")

-

+    

     meta_list <- metadata(rangesList)

     samples <- .check_metadata_list(metadata, metadata_prefix,meta_list)

     if(length(unlist(samples))<=0)

@@ -156,75 +173,92 @@ filter_and_extract <- function(data, metadata = NULL,

     }, meta_list, seq_along(meta_list))

 }

-.check_metadata_files <- function(metadata,metadata_prefix,meta_files)

+.check_metadata_files <- function(metadata,metadata_prefix,meta_files,col_name)

 {

+    suffix <- paste0(col_name,"$")

     vec_meta <- paste0(metadata_prefix,metadata)

     meta_list <- lapply(meta_files, function(x){

         list <- .add_metadata(x)

         vec_names <- names(list)

         a <- sapply(vec_meta, function(y) {

-            grep(y,vec_names) 

+            grep(y,vec_names)

         })

-

         ## we would like that manage more index from grep

         found <- as.logical(length(unlist(a)))

+        index <- grep(suffix,vec_names)

+        suffix <- unlist(list[index])[1] # ne prendo solo uno

+        names(suffix) <- NULL

         #if found retrieve samples that has at least one choosen metadata

-        if(found){x}

+        if(found)

+            list("sample" = x, "suffix" = suffix )

+        else

+            list("sample" = NULL, "suffix" = suffix )

+            

     })

 }

-.parse_gtf_files <- function(gtf_region_files,regions)

+.parse_gtf_files <- function(gtf_region_files, regions, suffixes)

 {

     g1 <- rtracklayer::import(con = gtf_region_files[1], format = "gtf")

     elementMetadata(g1) <- NULL

+    if(is.null(suffixes))

+        suffixes = ""

+    

     if(!is.null(regions))

     {

-        DF_list <- lapply(gtf_region_files, function(x){

+        DF_list <- mapply(function(x,h){

             g_x <- rtracklayer::import(con = x, format = "gtf")

             meta <- elementMetadata(g_x)[regions]

+            if(h!="")

+                names(meta) <- paste(regions,h,sep = ".")

             data.frame(meta)

-        })

+        },gtf_region_files, suffixes, SIMPLIFY = FALSE)

         DF_only_regions <- dplyr::bind_cols(DF_list)

         elementMetadata(g1) <- DF_only_regions

     }

     g1

 }

-.parse_gdm_files <- function(vector_field,gdm_region_files,regions)

+.parse_gdm_files <- function(vector_field,gdm_region_files,regions,suffixes)

 {

     #read first sample cause chromosome regions are the same for all samples

     df <- data.table::fread(gdm_region_files[1],col.names = vector_field,

-                                header = FALSE,sep = '\t')

+    header = FALSE,sep = '\t')

     col_names <- names(df)

-    df <-  df[c("chr","left","right","strand")]

-

+    df <-  subset(df, TRUE, c("chr","left","right","strand"))

+    

     if(!is.null(regions))

     {

         df_list <- lapply(gdm_region_files,function(x,regions,vector_field){

             region_frame <- data.table::fread(x,col.names = vector_field,

-                                                header = FALSE,sep = '\t')

+                    header = FALSE,sep = '\t')

             col_names <- names(region_frame)

             #delete column not choosen by input

             if(!is.null(regions))

-                col_names <- col_names[col_names %in% regions] 

-            r <- region_frame[col_names]

+                col_names <- col_names[col_names %in% regions]

+            

+            if(length(col_names)!=0)

+                r <- subset(region_frame,TRUE,col_names)

         },regions,vector_field)

-

         df_only_regions <- dplyr::bind_cols(df_list)

         complete_df <- dplyr::bind_cols(df,df_only_regions)

+        

+        region_names <- names(complete_df)[-(1:4)]

+        region_names <- gsub('[0-9]+', '',region_names)

+        region_names <- paste(region_names,suffixes,sep = ".")

+        region_names <- c(names(complete_df)[(1:4)],region_names )

+        names(complete_df) <- region_names

         g <- GenomicRanges::makeGRangesFromDataFrame(complete_df,

-                                            keep.extra.columns = TRUE,

-                                            start.field = "left",

-                                            end.field = "right")

+                            keep.extra.columns = TRUE,

+                            start.field = "left",

+                            end.field = "right")

     }

     else

-        g <- GenomicRanges::makeGRangesFromDataFrame(df,

-                                            keep.extra.columns = TRUE,

-                                            start.field = "left",

-                                            end.field = "right")

-

+        g <- GenomicRanges::makeGRangesFromDataFrame(df, 

+                            keep.extra.columns = TRUE,

+                            start.field = "left",

+                            end.field = "right")

+    

 }

-

-

@@ -87,7 +87,7 @@ check.DISTAL <- function(value)

 #' # and HM samples are obtained from the same provider (joinby clause).

 #' 

 #' join_data = merge(TSS, HM, 

-#' genometric_predicate = list(list(MD(1), DL(1200))), DF("provider"), 

+#' genometric_predicate = list(MD(1), DL(1200)), DF("provider"), 

 #' region_output = "RIGHT")

 #'

 #' #' # Given a dataset 'hm' and one called 'tss' with a sample including 

@@ -98,7 +98,7 @@ check.DISTAL <- function(value)

 #' # from the same provider (joinby clause).

 #' 

 #' join_data = merge(TSS, HM, 

-#' genometric_predicate = list(list(MD(1), DGE(12000), DOWN())), 

+#' genometric_predicate = list(MD(1), DGE(12000), DOWN()), 

 #' DF("provider"), region_output = "RIGHT")

 #'

 #' @name DL

@@ -23,7 +23,7 @@

     schema_name_xml <- list.files(datasetName, pattern = "*.xml$",

                                 full.names = TRUE)

-    if(length(schema_name)==0 || length(schema_name_xml) == 0)

+    if(length(schema_name)==0 && length(schema_name_xml) == 0)

         stop("schema not present")

     xml_schema <- xml2::read_xml(schema_name)

@@ -12,9 +12,8 @@

 #' only those samples that have the same value for each attribute

 #' are considered when performing the difference.

 #'

-#' @importFrom rJava J

-#' @importFrom rJava .jnull

-#' @importFrom rJava .jarray

+#' @importFrom rJava J .jnull .jarray

+#' @importFrom BiocGenerics setdiff

 #' 

 #' @param x GMQLDataset class object

 #' @param y GMQLDataset class object

new file mode 100644

@@ -0,0 +1,79 @@

+#' Method group_by

+#' 

+#' 

+#' @importFrom rJava J .jarray .jnull

+#' 

+#' 

+#' @param .data GMQLDataset object

+#'

+#' @return GMQLDataset object. It contains the value to use as input 

+#' for the subsequent GMQLDataset method

+#'

+#' @examples

+#' 

+#' 

+#' 

+#' @aliases group_by group_by-method

+#' @export

+setMethod("group_by","GMQLDataset",

+          function(.data, groupBy_meta = NULL, groupBy_regions = NULL, 

+            region_aggregates = NULL, meta_aggregates = NULL)

+          {

+              ptr_data = .data@value

+              gmql_group(ptr_data, groupBy_meta, groupBy_regions, 

+                            region_aggregates, meta_aggregates)

+          })

+

+gmql_group <- function(input_data, group_meta, group_reg)

+{

+    

+    if(!is.null(group_meta))

+    {

+        cond <- .join_condition(group_meta)

+        join_condition_matrix <- .jarray(cond, dispatch = TRUE)

+    }

+    else

+        join_condition_matrix <- .jnull("java/lang/String")

+    

+    if(!is.null(group_reg))

+    {

+        if(!is.character(group_reg))

+            stop("metadata: no valid input")

+        

+        group_reg <- group_reg[!group_reg %in% ""]

+        group_reg <- group_reg[!duplicated(group_reg)]

+        

+        if(length(group_reg)==0)

+            group_reg <- .jnull("java/lang/String")

+        

+        group_reg <- .jarray(metadata)

+    }

+    else

+        group_reg <- .jnull("java/lang/String")

+    

+    if(!is.null(meta_aggregates) && !length(meta_aggregates) == 0)

+    {

+        aggr <- .aggregates(meta_aggregates,"AGGREGATES")

+        metadata_matrix <- .jarray(aggr, dispatch = TRUE)

+    }

+    else

+        metadata_matrix <- .jnull("java/lang/String")

+    

+    if(!is.null(region_aggregates) && !length(region_aggregates) == 0)

+    {

+        aggr <- .aggregates(region_aggregates,"AGGREGATES")

+        region_matrix <- .jarray(aggr, dispatch = TRUE)

+    }

+    else

+        region_matrix <- .jnull("java/lang/String")

+    

+    WrappeR <- J("it/polimi/genomics/r/Wrapper")

+    response <- WrappeR$group(join_condition_matrix, metadata_matrix, 

+                                groupBy_regions, region_matrix, input_data)

+    error <- strtoi(response[1])

+    data <- response[2]

+    if(error!=0)

+        stop(data)

+    else

+        GMQLDataset(data)

+}

@@ -10,9 +10,8 @@

 #' The output metadata are the union of the input metadata, 

 #' with their attribute names prefixed with left or right respectively.

 #'

-#' @importFrom rJava .jnull

-#' @importFrom rJava J

-#' @importFrom rJava .jarray

+#' @importFrom rJava J .jnull .jarray

+#' @importFrom S4Vectors merge

 #' 

 #' @param x GMQLDataset class object

 #' @param y GMQLDataset class object

@@ -106,7 +105,7 @@ gmql_join <- function(left_data, right_data, genometric_predicate, joinBy,

         if(!all(sapply(genometric_predicate, function(x) {is(x,"DISTAL")} )))

             stop("All elements should be DISTAL object")

-         

+        

         genomatrix <- t(sapply(genometric_predicate, function(x) {

                 new_value = as.character(x)

                 array <- c(new_value)

@@ -39,7 +39,7 @@ execute <- function()

         {

             url <- WrappeR$get_url()

             .download_or_upload()

-            serialize_query(url,FALSE,data)

+            res <- serialize_query(url,FALSE,data)

         }

     }

 }

@@ -107,13 +107,18 @@ setMethod("collect", "GMQLDataset",

 gmql_materialize <- function(input_data, dir_out, name)

 {

-    dir_out <- sub("/*[/]$","",dir_out)

-    

-    res_dir_out <- paste0(dir_out,"/",name)

-    if(!dir.exists(res_dir_out))

-        dir.create(res_dir_out)

-    

     WrappeR <- J("it/polimi/genomics/r/Wrapper")

+    remote_proc <- WrappeR$is_remote_processing()

+    if(!remote_proc)

+    {

+        dir_out <- sub("/*[/]$","",dir_out)

+        res_dir_out <- paste0(dir_out,"/",name)

+        if(!dir.exists(res_dir_out))

+            dir.create(res_dir_out)

+    }

+    else

+        res_dir_out <- dir_out

+    

     response <- WrappeR$materialize(input_data, res_dir_out)

     error <- strtoi(response[1])

     data <- response[2]

@@ -10,9 +10,8 @@

 #' Samples whose names are not present in the grouping metadata parameter 

 #' are disregarded.

 #'

-#' @importFrom rJava J

-#' @importFrom rJava .jnull

-#' @importFrom rJava .jarray

+#' @importFrom rJava J .jarray .jnull

+#' @importFrom S4Vectors aggregate

 #'

 #' @param x GMQLDataset class object

 #' @param ... Additional arguments for use in specific methods.

@@ -244,7 +244,7 @@ We provide two metadata for you")

     {

         col_names <- plyr::revalue(col_names,c(start = "left", 

                                         end = "right", seqnames = "chr"))

-        schema_matrix <- cbind(toupper(col_types),col_names)

+        schema_matrix <- cbind(col_names,toupper(col_types))

         schema_matrix<- schema_matrix[setdiff(rownames(schema_matrix),

                                         c("group","width")),]

     }

@@ -10,10 +10,10 @@

 #' If no metadata in common between input dataset and semijoin dataset, 

 #' no sample is extracted.

 #'

-#' @importFrom rJava J

-#' @importFrom rJava .jnull

-#' @importFrom rJava .jarray

+#' @importFrom rJava J .jnull .jarray

 #' @importFrom methods isClass

+#' @importFrom glue glue

+#' @importFrom dplyr filter

 #' 

 #' @param .data GMQLDataset class object

 #' @param m_predicate logical predicate made up by R logical operation 

@@ -69,31 +69,34 @@

 #' @aliases filter filter-method

 #' @export

 setMethod("filter", "GMQLDataset",

-            function(.data, m_predicate = NULL, r_predicate = NULL, 

-                        semijoin = NULL, ...)

+        function(.data, m_predicate = NULL, r_predicate = NULL, 

+                    semijoin = NULL, ...)

+        {

+            val <- .data@value

+            meta_pred <- substitute(m_predicate)

+            if(!is.null(meta_pred))

             {

-                val <- .data@value

-                meta_pred <- substitute(m_predicate)

-                if(!is.null(meta_pred))

-                {

-                    predicate <- .trasform(deparse(meta_pred))

-                    predicate <- paste(predicate,collapse = "")

-                }

-                else

-                    predicate <- .jnull("java/lang/String")

+                predicate <- .trasform(deparse(meta_pred))

+                predicate <- paste(predicate,collapse = "")

+                predicate <- as.character(glue::glue(predicate))

-                reg_pred <- substitute(r_predicate)

-                if(!is.null(reg_pred))

-                {

-                    region_predicate <- .trasform(deparse(reg_pred))

-                    region_predicate <- paste(region_predicate,collapse = "")

-                }

-                else

-                    region_predicate <- .jnull("java/lang/String")

-

-                gmql_select(val, predicate, region_predicate, semijoin)

-            })

-

+            }

+            else

+                predicate <- .jnull("java/lang/String")

+            

+            reg_pred <- substitute(r_predicate)

+            if(!is.null(reg_pred))

+            {

+                region_predicate <- .trasform(deparse(reg_pred))

+                region_predicate <- paste(region_predicate,collapse = "")

+                region_predicate <- as.character(glue::glue(region_predicate))

+                

+            }

+            else

+                region_predicate <- .jnull("java/lang/String")

+            

+            gmql_select(val, predicate, region_predicate, semijoin)

+        })

 gmql_select <- function(input_data, predicate, region_predicate, s_join)

 {

@@ -22,6 +22,7 @@

 #' the dataset to which they originally belonged.

 #' 

 #' @importFrom rJava J

+#' @importFrom BiocGenerics union

 #' 

 #' @param x GMQLDataset object

 #' @param y GMQLDataset object 

@@ -1052,12 +1052,14 @@ sample_metadata <- function(url, datasetName,sampleName)

 #'

 sample_region <- function(url, datasetName,sampleName)

 {

+    

     url <- sub("/*[/]$","",url)

     URL <- paste0(url,"/datasets/",datasetName,"/",sampleName,"/region")

     h <- c('X-Auth-Token' = authToken, 'Accpet' = 'text/plain')

     req <- httr::GET(URL, httr::add_headers(h))

     content <- httr::content(req, 'parsed',encoding = "UTF-8")

+    

     if(req$status_code !=200)

         stop(content)

     else

@@ -1065,6 +1067,8 @@ sample_region <- function(url, datasetName,sampleName)

         list <- show_schema(url,datasetName)

         schema_type <- list$type

+        #df <- read.table(textConnection(content),sep = "\t")

+        

         temp <- tempfile("temp") #use temporary files

         write.table(content,temp,quote = FALSE,sep = '\t',col.names = FALSE,

                         row.names = FALSE)

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

 ##gff-version 2

 ##source-version rtracklayer 1.36.6

-##date 2017-11-19

+##date 2017-12-04

 chr1	rtracklayer	sequence_feature	11873	12227	0	+	.	name "NR_046018"

 chr1	rtracklayer	sequence_feature	12612	12721	0	+	.	name "NR_046018"

 chr1	rtracklayer	sequence_feature	917444	917497	0	-	.	name "NM_001291366"

Binary files a/inst/java/GMQL.jar and b/inst/java/GMQL.jar differ

@@ -1,14 +1,11 @@

 % Generated by roxygen2: do not edit by hand

-% Please edit documentation in R/AllGenerics.R, R/gmql_merge.R

+% Please edit documentation in R/gmql_merge.R

 \docType{methods}

-\name{aggregate}

-\alias{aggregate}

+\name{aggregate,GMQLDataset-method}

 \alias{aggregate,GMQLDataset-method}

 \alias{aggregate-method}

 \title{Method aggregate}

 \usage{

-aggregate(x, ...)

-

 \S4method{aggregate}{GMQLDataset}(x, ...)

 }

 \arguments{

@@ -31,8 +28,6 @@ GMQLDataset object. It contains the value to use as input

 for the subsequent GMQLDataset method

 }

 \description{

-Wrapper to GMQL merge function

-

 It builds a dataset consisting of a single sample having as many regions as

 the number of regions of the input data and as many metadata as the union of

 the 'attribute-value' tuples of the input samples. 

@@ -85,7 +85,7 @@ HM = read_dataset(test_path2)

 # and HM samples are obtained from the same provider (joinby clause).

 join_data = merge(TSS, HM, 

-genometric_predicate = list(list(MD(1), DL(1200))), DF("provider"), 

+genometric_predicate = list(MD(1), DL(1200)), DF("provider"), 

 region_output = "RIGHT")

 #' # Given a dataset 'hm' and one called 'tss' with a sample including 

@@ -96,7 +96,7 @@ region_output = "RIGHT")

 # from the same provider (joinby clause).

 join_data = merge(TSS, HM, 

-genometric_predicate = list(list(MD(1), DGE(12000), DOWN())), 

+genometric_predicate = list(MD(1), DGE(12000), DOWN()), 

 DF("provider"), region_output = "RIGHT")

 }

@@ -78,7 +78,7 @@ test_path2 <- system.file("example", "DATASET_GDM", package = "RGMQL")

 data <- read_dataset(test_path)

 join_data <- read_dataset(test_path2)

 jun_tf <- filter(data, antibody_target == "JUN", pValue < 0.01, 

-semijoin(join_data, TRUE, DF("cell")))

+semijoin(join_data, TRUE, list(DF("cell"))))

 }

@@ -5,42 +5,49 @@

 \title{FILTER AND EXTRACT}

 \usage{

 filter_and_extract(data, metadata = NULL, metadata_prefix = NULL,

-  regions = NULL)

+  regions = NULL, suffix = "antibody_target")

 }

 \arguments{

-\item{data}{string GMQL dataset folder path or GrangesList 

+\item{data}{string GMQL dataset folder path or GrangesList

 object}

-\item{metadata}{vector of character containing names of metadata 

+\item{metadata}{vector of character containing names of metadata

 to be searched for in metadata files.

 data will be extracted if at least one condition is satisfied:

 this condition will be logically "ANDed" with prefix filtering (see below)

-if NULL no filtering action occured 

+if NULL no filtering action occured

 (i.e every sample will be taken for regions filtering)}

-\item{metadata_prefix}{vector of character that will filter metadata 

+\item{metadata_prefix}{vector of character that will filter metadata

 containing rispectively every element of this vector.

 number of elelment in both vector must match}

-\item{regions}{vector of character that will extract only region 

-attribute specified; if NULL no regions will be taken and the output 

+\item{regions}{vector of character that will extract only region

+attribute specified; if NULL no regions will be taken and the output

 will be only GRanges made up by the first attribute

 (seqnames,start,end,strand)}

+

+\item{suffix}{name for each metadata column of GRanges. by default is the 

+"antibody_target". This string is taken from sample metadata file or from

+metadata() associated. If not present, the column name is the name of 

+selected regions}

 }

 \value{

 granges with selected regions (if any) in elementMetadata

 }

 \description{

-This function let user to create a new GRangeslist with fixed information: 

-seqnames,ranges ans strand and a variable part made up by the regions 

+This function let user to create a new GRangeslist with fixed information:

+seqnames,ranges ans strand and a variable part made up by the regions

 defined as input.

-The metadata and metadata_prefix are used to filter the data and choose 

+The metadata and metadata_prefix are used to filter the data and choose

 only the samples that match at least one metdatata with its prefix.

 The input regions are shown for each sample obtained from filtering.

 }

 \details{

-This function works only with datatset or Grangeslist that have the same 

+This function works only with datatset or Grangeslist that have the same

 information about regions attribute (but of course different value)

+In case of Grangeslist data input the function will search for metadata

+into metadata() function associated to Grangeslist.

 }

 \examples{

new file mode 100644

@@ -0,0 +1,31 @@

+% Generated by roxygen2: do not edit by hand

+% Please edit documentation in R/AllGenerics.R, R/gmql_group.R

+\docType{methods}

+\name{group_by}

+\alias{group_by}

+\alias{group_by,GMQLDataset-method}

+\alias{group_by}

+\alias{group_by-method}

+\title{Method group_by}

+\usage{

+group_by(.data, ...)

+

+\S4method{group_by}{GMQLDataset}(.data, ...)

+}

+\arguments{

+\item{.data}{GMQLDataset object}

+}

+\value{

+GMQLDataset object. It contains the value to use as input 

+for the subsequent GMQLDataset method

+}

+\description{

+Wrapper to GMQL group function

+

+Method group_by

+}

+\examples{

+

+

+

+}

@@ -51,6 +51,7 @@ test_path <- system.file("example", "DATASET", package = "RGMQL")

 test_path2 <- system.file("example", "DATASET_GDM", package = "RGMQL")

 data <- read_dataset(test_path)

 join_data <-  read_dataset(test_path2)

-jun_tf <- filter(data,NULL,NULL, semijoin(join_data, TRUE, DF("cell")))

+jun_tf <- filter(data,NULL,NULL, semijoin(join_data, TRUE, 

+list(DF("cell"))))

 }

@@ -390,7 +390,7 @@ Upon succesfull logon, you get a request token that you must use

 in every subsequent REST call.

 Login can be performed using the function:

 ```{r, eval = TRUE}

-test_url = "http://genomic.elet.polimi.it/gmql-rest-r"

+test_url = "http://genomic.elet.polimi.it/gmql-rest-r/"

 login_gmql(test_url)

 ```

 that saves the token in the Global R environment within the variable 

@@ -436,6 +436,12 @@ name_dataset <- job$datasets[[1]]$name

 download_dataset(test_url,name_dataset)

 ```

+Once done we can loogut from remote repository using: