I have 17 samples collected through patch-seq, processed using SMART-Seq HT Plus Kit, and sequenced on the HiSeq X platform. I used STAR v2.7.7a to align my reads to the mouse genome mm10 (my reference genome). I'm relatively new to this field and would like to understand the difference between unmapped reads (flagged by STAR) and duplicate reads (reported by FastQC). We have a high value of duplicate reads which I read is "normal" for single-cell RNA seq (assuming that it applies to patch-seq too) because of overamplification which is inevitable for a small amount of starting material. In addition to this, we have some samples with high unmapped reads too. Are the unmapped reads the same as duplicate reads? Or do they include duplicate reads? I have attached a link for the graph of %unmapped reads and %duplicate reads for all my samples.
Thanks in advance.