I'm using STAR to analyse RNA-seq data. I used GRCh38 as reference genome and also the annotation file from ensembl site under same build. The result from STAR shows a high percentage no_feature alignments which are those aligned to reference genome but not annotated. I know 18S and 28S are not annotated in the gtf file and I wonder if the reference genome contain those sequences. Because if it contains, then the no_feature may be due to failed rRNA depletion.