our workflow (we usually get multiplexed fastq files): mapping (bowtie1) to PhiX => demultiplexing => quality filtering with fastx => mapping to rDNA => mapping to ref Organism => sorting and filtering (

samtools view -b -q 20

) bam files and converting them to bed => convert bed to ctss by splitting them into strands and grouping by chr name and start coordinate and putting the count into column 6 => pooling all ctss files togeter => sorting and grouping/counting the pooled ctss => Filter away rows in ctssall file with TPM<0.1 => splitting into strands and putting it into paraclu specific format => run paraclu clustering on each strand => run paraclu_cut on each strand => merge strands and convert back into bed6 file => intersect individual ctss files with paraclu cluster coordinates