I am using STAR to analyze a small RNA-seq dataset for the first time. There were already some issues with depth so there's just about 5M reads per sample. I used ENCODE's small RNA parameters:
params=' --runThreadN 4
--quantMode TranscriptomeSAM GeneCounts
--outSAMtype BAM SortedByCoordinate
There are two major concerns I have from the alignment. These are a few red flags from the log file:
Uniquely mapped reads number | 614121
Uniquely mapped reads % | 22.03%
Average mapped length | 50.49
There is a low number of uniquely mapped reads and the mapped lengths are too long to be mirna.
Furthermore, when running featureCounts, none of my aligned reads were assigned to a mature miRNAs from the gff3 file on mirbase.
Do you think these issues come from low depth and poor size selection?
Thank you in advance tor your assistance!