dtriple negative fragment sizes with paird-end input

[Talon Trecek]

Hello,

I have two replicates for K27me3 generated by CUT&RUN sequenced by paired-end.

From the dtriple output, the most enriched fragment size on the first rep is -179 and the second rep 179.

Is this negative fragment size expected?

I don't see any blatant affect on output.

Both reps have enrichment at -179, but rep2 appears to be a bit higher at 179.

I suspect this is an issue with how the input is formatted.

The manual specifies that paired end input bed files must have mates in neighboring lines. I have sorted the input bed files by read name (samtools sort -n | bamtobed). The read names follow standard naming with the only difference between mates being /1 /2.

Perhaps the problem is that the strand is not accounted for when subtracting the start of one mate with the end of the other?

thanks,

Talon