upgrade part1

R/build_genome_features.R

@@ -72,8 +72,12 @@ setMethod("build_genome_features",definition=function(txdb,features,parameters)
   ###################
   # build ids cross reference
   crossref<-data.table::data.table(
-    AnnotationDbi::select(txdb,columns=c("TXID","TXNAME"),
-    keytype="GENEID",keys=AnnotationDbi::keys(txdb))
+    AnnotationDbi::select(
+      txdb,
+      columns=c("TXID","TXNAME"),
+      keytype="GENEID",
+      keys=AnnotationDbi::keys(txdb)
+    )
   )
 
   ###################
@@ -387,58 +391,76 @@ setMethod("build_genome_features",definition=function(txdb,features,parameters)
   # load annotation supplement
   ###################
 
-  ###################
-  # read the gff
-  file=AnnotationDbi::metadata(txdb)$value[3]
-
-  if(!base::is.na(file)){
-
     ###################
-    # download
-    if(base::length(grep("ftp.+\\.gz$",file))>0){
+    # check file name
+    file=AnnotationDbi::metadata(txdb)$value[3]
+
+    if(!base::is.na(file)){
 
       ###################
-      # import the GFF
-      utils::download.file(file,quiet=T,destfile = "gff.gz")
+      # download
+      if(base::length(grep("ftp.+\\.gz$",file))>0){
 
-      ##################
-      #  uncompress
-      R.utils::gunzip("gff.gz")
+        ###################
+        #  output name
+        output=base::sub("^.+/","",file)
 
-      ##################
-     # file to download
-      file="gff"
-    }
+        ###################
+        # import the GFF
+        utils::download.file(file,quiet=T,destfile=output)
 
-    ###################
-    # read the gff
-    #gff<-data.table::fread(file,fill=T,drop=base::c(2,6,8),verbose=F,showProgress=F,col.names=base::c("chr",
-    #"type","start","end","strand","att"))[data.table::like(type,"RNA|transcript") & !data.table::like(chr,"^#")]
+        ##################
+        #  uncompress
+        R.utils::gunzip(output)
 
-    gff<-data.table::fread(file,sep="",verbose=F,showProgress=F,header=F)[data.table::like(base::gsub("ID=.+$","",V1),"[[:space:]].+(RNA|transcript)"),]
+        ##################
+        # file to download
+        file=output
+      }
 
-    ###################
-    # parse it
-    gff<-gff[,{
-      tmp=data.table::data.table(base::t(base::strsplit(V1,"\t")[[1]]));
-      base::list(chr=tmp$V1,type=tmp$V3,start=tmp$V4,end=tmp$V5,strand=tmp$V7,att=tmp$V9)
-      },by=1:base::nrow(gff)][,base:=NULL]
-
-    ###################
-    # remove files
-    if(file=="gff"){base::unlink("gff");base::unlink("gff.gz")}
+      ###################
+      # read the gene file as GRanges
+      gene_table<-rtracklayer::import(file)
 
-    ###################
-    # extract all GeneID
-    gff<-GenomeFeatures::GFF_att(gff,att.list = c("Name","GeneID","gene"))
+      ###################
+      # convert to data.table
+      gene_table<-data.table::data.table(
+        BiocGenerics::as.data.frame(gene_table,,row.names = NULL)
+      )[ data.table::like(type,"RNA|transcript"),]
+
+
+      if(base::length(base::grep("gtf$",file))>0){
+
+        ###################
+        # select columns from gtf
+        gene_table<-gene_table[,
+          .(seqnames,start,end,strand,type,gene_id,gene_name,gene_biotype,transcript_id)
+        ]
+
+      }else{
+
+        ###################
+        # select columns from gff
+        gene_table<-gene_table[,
+          .(seqnames,start,end,strand,type,Dbxref,gene,transcript_id)
+        ]
+
+        ###################
+        # extract gene_id
+        gene_table[,
+          `:=`(
+          gene_id=base::sub("^.+:","",Dbxref[[1]][1]),
+          Dbxref=NULL
+          ),
+          by=1:base::nrow(gene_table)
+        ]
+      }
 
-    ###################
-    # extract all GeneID
-    gff<-base::unique(gff[,.(type,Name,GeneID,gene)])
   }else{
+
     ###################
     # else NA
-    gff=data.table::data.table(gff=NA)
+    gene_table=data.table::data.table(gene_table=NA)
   }
 
   ###################
@@ -456,6 +478,6 @@ setMethod("build_genome_features",definition=function(txdb,features,parameters)
     intron=base::list(GRanges=introns,DT=introns_DT),
     cds=base::list(GRanges=cds,DT=cds_DT),
     downstream=base::list(GRanges=downstream,DT=downstream_DT),
-    gff=gff
+    gene_table=gene_table
   )
 })

R/genome_features.R

@@ -24,7 +24,7 @@ setClass("genome_features",
            cds="list",
            UTR3="list",
            downstream="list",
-           gff="data.table"
+           gene_table="data.table"
          )
 )
 
@@ -60,19 +60,19 @@ setMethod("show", "genome_features",function(object) {
 
   ###################
   # extract gff
-  gff<-methods::slot(object,"gff")
+  gene_table<-methods::slot(object,"gene_table")
 
   ###################
   # count genes
-  genes<-base::length(base::unique(gff$GeneID))
+  genes<-base::length(base::unique(gene_table$GeneID))
 
   ###################
   # count
-  gff<-gff[,.N,by=type]
+  gene_table<-gene_table[,.N,by=type]
 
   ###################
   # chaneg header
-  base::names(gff)<-base::c("feature","count")
+  base::names(gene_table)<-base::c("feature","count")
 
   ###################
   # bind results
@@ -80,7 +80,7 @@ setMethod("show", "genome_features",function(object) {
     base::c(
       base::list(data.table::data.table(feature="gene",count=genes)),
       Data,
-      base::list(gff)
+      base::list(gene_table)
     )
   )
 

R/genome_features_overlaps.R

@@ -219,11 +219,11 @@ setMethod("genome_features_overlaps",definition=function(peaks,features,figs_pat
     # remove tag
     annot[,tag:=NULL]
 
-    if(base::ncol(features@gff)>1){
+    if(base::ncol(features@gene_table)>1){
 
       ###################
       # merge with overlapped features
-      annot<-merge(annot,features@gff,by.x="tx_name",by.y="Name",all.x=T,sort=F)
+      annot<-merge(annot,features@gene_table,by.x="tx_name",by.y="Name",all.x=T,sort=F)
     }else{
 
       ###################

R/overlapped_genome_features.R

@@ -29,21 +29,16 @@ setMethod("show", "overlapped_genome_features",function(object) {
   # extract annotations
   Data<-methods::slot(object,"annotation")
 
-  ###################
-  # available columns
-  colnum=base::c("condition","feature_type","type","GeneID")
-  colnum= colnum[colnum%in%base::names(Data)]
-
   ###################
   # count unique genes match by feature type
-  Data<-base::unique(Data[,.SD,.SDcols=colnum])
+  Data<-base::unique(Data[,.(condition,feature_type,type,gene_id)])
 
   ###################
   # count
   Data<-data.table::dcast(
     Data,
     condition~feature_type,
-    value.var="GeneID",
+    value.var="gene_id",
     fun=length
   )