Help, How do I add TF information to a Peak-to-Gene DataFrame?

[davidlyon3]

---

How do I add TF information to a Peak-to-Gene DataFrame?

I have the following DataFrame, p2geneDF, which contains information about peaks and their associated genes. I want to add columns to this DataFrame that include transcription factor (TF) information, specifically linking each peak to the TFs that bind to it. This will allow me to identify both the TF and the genes being regulated by it.

Thank you very much in advance.

p2geneDF <- metadata(proj@peakSet)$Peak2GeneLinks
p2geneDF$geneName <- mcols(metadata(p2geneDF)$geneSet)$name[p2geneDF$idxRNA]
p2geneDF$peakName <- (metadata(p2geneDF)$peakSet %>% {paste0(seqnames(.), "_", start(.), "_", end(.))})[p2geneDF$idxATAC]

> p2geneDF
DataFrame with 153894 rows and 8 columns
         idxATAC    idxRNA         Correlation                  FDR
       <integer> <integer>           <numeric>            <numeric>
1              1         2 0.00714436960966568    0.891975576529501
2              2         2   0.206067895564052 1.08749544612577e-05
3              3         2   0.187709745154478  6.6277728575822e-05
4              4         2   0.230970917550565 7.09785052131589e-07
5              5         2   0.232274447292216 6.09708826430438e-07
...          ...       ...                 ...                  ...
153890     25504      3078   0.315701130804448 4.88826759376223e-12
153891     25505      3078   0.519923106200753 1.64848166416267e-33
153892     25506      3078   0.375368854528432 7.73492476815824e-17
153893     25507      3078  -0.444076195471676 9.17186406922461e-24
153894     25508      3078  -0.124051315528399  0.00960714755007798
                VarQATAC           VarQRNA  geneName                 peakName
               <numeric>         <numeric>   <array>              <character>
1      0.185941665359887 0.446718648473034 LINC00115       chr1_757871_758371
2       0.44766347812451 0.446718648473034 LINC00115       chr1_758489_758989
3      0.432099733416967 0.446718648473034 LINC00115       chr1_761835_762335
4      0.860906382311432 0.446718648473034 LINC00115       chr1_762697_763197
5      0.226556374470754 0.446718648473034 LINC00115       chr1_764481_764981
...                  ...               ...       ...                      ...
153890 0.128312686216089 0.423001949317739     PDCD1 chr2_243029037_243029537
153891 0.953151952328681 0.423001949317739     PDCD1 chr2_243030546_243031046
153892 0.217186764936491 0.423001949317739     PDCD1 chr2_243031340_243031840
153893 0.869923161361142 0.423001949317739     PDCD1 chr2_243040594_243041094
153894  0.23906225497883 0.423001949317739     PDCD1 chr2_243050293_243050793

I also have TF binding site information stored in motif_matches (RangedSummarizedExperiment object) or similar data, which links peaks to TF motifs. How can I use this to enrich p2geneDF with the TF information (e.g., a column listing the TFs associated with each peak)?

My goal is to:

1. Add one or more columns to p2geneDF that include the TFs associated with each peak.
2. Ensure that I can identify the TFs regulating each gene via the peak-to-gene links.

If there are existing methods or best practices for this in ArchR, I'd appreciate guidance! Thank you.