I'm working with RNA-seq data and performed transcriptome assembly using StringTie, followed by classification with GffCompare. I'm observing something puzzling: some transcripts that appear to originate from the same gene locus are being classified differently — some as x (antisense) and others as u (intergenic). From my understanding, 'x' indicates overlap with a known gene on the opposite strand, while 'u' means the transcript is intergenic. So, I was wondering:

• Is it biologically or technically possible for the same gene locus to produce both 'u' and 'x' transcripts?
• Could this be an artifact of transcript assembly or a reflection of complex transcriptional regulation?