This directory contains BED files of merged Transcription Start Sites (TSSs) generated within the GENCODE Capture Long-Seq project using PacBio HCGM reads, and split by known/novel status w.r.t GENCODE 20 (human) GENCODE M3 (mouse) .

These were produced by comparing *All_Cap1_all_TSSs.clusters.bed files listed in ../ (parent directory) to GENCODE TSSs using the following command:

$ cat <CLUST_TSS_BEDfile> | sortbed | bedtools closest -s -t first -D b -a stdin -b <GENCODE TSSs w/o chrM> | awk '$7!="."' | awk '$NF>=-50 && $NF<=50' | cut -f1-6 > <CLUST_TSS_BEDfile>.known.bed
$ cat <CLUST_TSS_BEDfile> | sortbed | bedtools closest -s -t first -D b -a stdin -b <GENCODE TSSs w/o chrM> | awk '$7!="."' | awk '$NF<-50 || $NF>50' | cut -f1-6 > <CLUST_TSS_BEDfile>.novel.bed


All files correspond to genome assemblies hg38 and mm10.


# File naming scheme:

<species>All_Cap1_<tissue>_TSSs.clusters.<gencode_status>.bed

where:
	species:
		"mm": mouse
		"hs": human

	tissue: 
		"all": all TSSs merged across all tissues.

	gencode_status:
		"known": CLS TSSs <= 50 bases away from a GENCODE TSS on the same genomic strand
		"novel": CLS TSSs > 50 bases away from a GENCODE TSS on the same genomic strand

# BED file format (BED6):

There is one read per BED record.

    column 1: chromosome 
    column 2: chromosome start of merged TSS
    column 3: chromosome end of merged TSS
    column 4: comma-separated list of read identifiers contributing to the TSS
    column 5: number of reads contributing to the TSS
    column 6: genomic strand of the TSS