How to interpret the results from LinkPeaks() analysis?

[AndiL04]

Hi,

I have recently been utilizing Signac for analysis on a 10x Multiome dataset. During this process, I successfully executed the LinkPeaks function on genes of interest.

However, I found difficulties to interprete some part of the results.

1. As illustrated in the accompanying screenshot, there appears to be a discrepancy between the significantly correlated genomic regions and the identified peaks. My initial assumption was that these genomic regions would represent the Transcription Start Sites (TSS) of the targeted genes.
2. Additionally, I am unclear why the linked genomic regions vary in width.

Thank you for your help!

Andi

[timoast]

The genomic ranges themselves (seqnames, start, end) give the coordinates of the link to be drawn when creating a plot, which is the midpoint of the peak and the TSS of the gene. The more informative part for an analysis is in the metadata columns: score, gene, peak, zscore, pvalue. These give the name of the gene and coordinates of the peak and the strength of the relationship. Please see the documentation for LinkPeaks for more

[AndiL04]

Thank you so much for the explanation!