I have small RNA sequencing data (fastQC read length graph below) from rat tissue across two conditions and would like to investigate the presence of microbiome-sourced RNA. I have added the read length graph from an example fastQC report.

I appreciate that this would significantly limit the phylogenetic resolution for Kraken2/Bracken, but I would hope that I can do some analysis on this. I have done a trial run following the Lu et al. (2022), with the "Standard" database (third one down at https://benlangmead.github.io/aws-indexes/k2) using the Kraken2 and Bracken commands-

kraken2 --db k2protocol_db --threads 8 --report Logs/${sample}.k2report --report-minimizer-data --minimum-hit-groups 3 Samples/${sample}.fq

./bracken -d ../k2protocol_db -i ../Logs/${sample}.k2report -r 100 -l F -t 10 -o ../bracken_outputs/${sample}_Family.bracken -w ../breports/${sample}_Family.breport

I acknowledge that a read length of 100 isn't applicable here (I still got lots of returned hits though?). I have read bits and pieces saying that it is not as simple as reducing -r to 21 for example. Can anyone advise on the needed adjustments to get any type of meaningful data from these reads?

Thanks