@@ -92,15 +92,30 @@ import_gmql <- function(dataset_path, is_gtf)

         name_samples <- lapply(regions, function(x){

             gsub("*.gdm", "",basename(x))})

         vector_field <- .schema_header(datasetName)

-

+        type_and_coord <- .schema_type_coordinate(datasetName)

         names(vector_field) <- NULL

-        sampleList <- lapply(regions,function(x){

-            df <- read.delim(x,col.names = vector_field,header = FALSE)

-            g <- GenomicRanges::makeGRangesFromDataFrame(df,

-                                            keep.extra.columns = TRUE,

-                                            start.field = "left",

-                                            end.field = "right")

-        })

+        if(type_and_coord$coordinate_system %in% c("1-based"))

+        {

+            sampleList <- lapply(regions,function(x){

+                df <- read.delim(x,col.names = vector_field,header = FALSE)

+                g <- GenomicRanges::makeGRangesFromDataFrame(df,

+                        keep.extra.columns = TRUE,

+                        start.field = "left",

+                        end.field = "right")

+            })

+        }

+        else

+        {

+            sampleList <- lapply(regions,function(x){

+                df <- read.delim(x,col.names = vector_field,header = FALSE)

+                df$left = df$left +1

+                g <- GenomicRanges::makeGRangesFromDataFrame(df,

+                        keep.extra.columns = TRUE,

+                        start.field = "left",

+                        end.field = "right")

+            })

+        }

+       

         names(sampleList) <- name_samples

         gRange_list <- GenomicRanges::GRangesList(sampleList)

     }

@@ -121,6 +121,7 @@ export_gmql <- function(samples, dir_out, is_gtf)

         lapply(samples,function(x,dir){

             sample_name <- file.path(dir,paste0("S_",cnt(),".gdm"))

             region_frame <- data.frame(x)

+            region_frame$start = region_frame$start - 1

             write.table(region_frame,sample_name,col.names = FALSE,

                             row.names = FALSE, sep = '\t',quote = FALSE)

         },files_sub_dir)

@@ -183,6 +184,17 @@ export_gmql <- function(samples, dir_out, is_gtf)

     xml2::xml_attr(root,"name") <- "DatasetName_SCHEMAS"

     xml2::xml_attr(root,"xmlns") <- "http://genomic.elet.polimi.it/entities"

     xml2::xml_add_child(root,"gmqlSchema")

+    if(to_GTF)

+    {

+        xml2::xml_attr(root,"type") <- "gtf"

+        xml2::xml_attr(root,"coordinate_system") <- "1-based"

+    }

+    else

+    {

+        xml2::xml_attr(root,"type") <- "tab"

+        xml2::xml_attr(root,"coordinate_system") <- "0-based"

+    }

+    

     gmqlSchema <- xml2::xml_child(root,1)

     names_node <- names(node_list)

@@ -36,6 +36,27 @@

     vector_field <- unlist(list_field)

 }

+.schema_type_coordinate <- function(datasetName)

+{

+    schema_name <- list.files(datasetName, pattern = "*.schema$",

+                              full.names = TRUE)

+    

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

+                                  full.names = TRUE)

+    

+    if(!length(schema_name) && !length(schema_name_xml))

+        stop("schema not present")

+    

+    if(!length(schema_name))

+        xml_schema <- xml2::read_xml(schema_name_xml)

+    else

+        xml_schema <- xml2::read_xml(schema_name)

+    

+    gmql_schema_tag <- xml2::xml_children(xml_schema)

+    all_attrs <- xml2::xml_attrs(gmql_schema_tag)

+    all_attrs_list <- as.list(all_attrs[[1]])

+}

+

 # aggregates factory

 .aggregates <- function(meta_data,class)

 {

@@ -11,7 +11,8 @@

 #' \item{TAB: tab-delimited file format}

 #' \item{GTF: tab-delimited text standard format based on the General 

 #' Feature Format}

-#' \item{COLLECT: used for storing output in memory}

+#' \item{COLLECT: used for storing output in memory (only in the case of local 

+#' processing, i.e., remote_processing = FALSE,)}

 #' }

 #' @param remote_processing logical value specifying the processing mode.

 #' True for processing on cluster (remote), false for local processing.

@@ -22,8 +23,8 @@

 #' You can always perform it by calling the function \code{\link{login_gmql}} 

 #' explicitly

 #' 

-#' @param username string name used during signup 

-#' @param password string password used during signup

+#' @param username string name used during remote server signup 

+#' @param password string password used during remote server signup

 #' 

 #' @return None

 #'

@@ -152,7 +152,7 @@ gmql_materialize <- function(input_data, dir_out, name)

         res_dir_out <- dir_out

     if(grepl("\\.",name))

-        stop("name dataset cannot contains dot")

+        stop("dataset name cannot contains dot")

     response <- WrappeR$materialize(input_data, res_dir_out)

     error <- strtoi(response[1])

@@ -753,14 +753,15 @@ By default *suffix* correspond to a metadata: *antibody_target*.

 ## Metadata

 Each sample of a GMQL dataset has its own metadata associated and generally 

-every metadata attribute has a single value.The case with distinct values 

-for the same metadata attribute, is showed in the figure below for the disease 

-metadata attribute 

+every metadata attribute has a single value. The case with distinct values 

+for the same metadata attribute is shown in the figure below for the 

+\emph{disease} metadata attribute.

 \newline\newline

 ![metadata with multiple values](multi_metadata.png)

 \newline\newline

-In this case GMQL automcatically handles this situation.

-Import/export paragraph we showed that a GMQL dataset can be imported into R environment as a GRangesList, and so its metadata too.

+In this case GMQL automatically handles this situation. In the Import/export 

+paragraph, we showed that a GMQL dataset can be imported into R environment 

+as a GRangesList, and so its metadata too.

 ```{r, metadata}

 # This statement defines the path to the folder "DATASET_META" in the 

@@ -781,8 +782,9 @@ metadata(grl_data)

 The metadata are stored as simple list in the form key-values and it does not 

 matter if mutiple values for the same metadata attribute are present; 

 all values are stored and shown.

-Difficulties can arise when we need to get all the metadata values; normally 

-since list is in the form key-value, we can extract the metadata values using:

+Difficulties can arise when we need to get all the metadata values; normally, 

+since the metadata list is in the form key-value, we can extract the metadata 

+values using:

 ```{r, retrieve_value}

@@ -800,7 +802,7 @@ all disease values, we should use instead:

 ```{r, retrieve_values}

-# get all disease value of sample S_00000

+# get all disease values of sample S_00000

 a$S_00000[which(names(a$S_00000) %in% "disease")]