Problem running with cage_peak about sqanti2

[Magdoll]

Hi @PJ-Shaw ,
SQANTI2 is using the format used by FANTOM5 for their CAGE Peak Bed. See data here. Additional info on the refTSS is in their publication

The required columns are

column 1 - chromosome
column 2 - cage peak signal start
column 3 - cage peak signal end
column 6 - strand
column 7 - cage peak "tss" <--- this is what FANTOM5 considers as the actual TSS

Example below

chr1    629886  629898  hg_3.1  1       +       629892  629893  255,255,0
chr1    629904  629938  hg_223433.1     1       +       629921  629922  255,255,0
chr1    629945  629967  hg_223434.1     1       -       629956  629957  255,255,0
chr1    630059  630067  hg_223435.1     1       +       630063  630064  255,255,0
chr1    630098  630103  hg_4.1  1       +       630100  630101  255,255,0
chr1    630129  630161  hg_5.1  1       +       630145  630146  255,255,0
chr1    630225  630237  hg_236023.1     1       +       630231  630232  255,255,0

[PJ-Shaw]

Thank you for spotting the error. I added a 7th column with the "dominant_ctss" nucleotide reported by CAGEr to my bed file and SQANTI2 worked, reporting isoforms "within_cage_peak".