You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
After running the pipeline I noticed that while the FIMO matrix returns ~1,000 TFs and the gene-TF links return ~800 TFs, the In silico CHIP chromVAR scores are only calculated for ~200 TFs.
The text was updated successfully, but these errors were encountered:
You can adjust the parameters used to perform filtering in config. yaml. The default is requiring 30 peaks with an in silico ChIP score > .15 for a TF to pass filtering:
ins_chip:
# set verbose to "--verbose" for True
verbose: ""
min_chip_score: 0.15
min_peak_hits: 30
The NAs in the chromVAR deviations occur when the expected number of fragments in the cell is less than 1, according to GreenleafLab/chromVAR/issues/27. I've been filling the NA values with the minimum chromVAR score for that TF. Signac fills the missing values with zeros as seen in stuart-lab/signac/issues/1254. Missing values need to be filled in run_chromvar.R before export.
After running the pipeline I noticed that while the FIMO matrix returns ~1,000 TFs and the gene-TF links return ~800 TFs, the In silico CHIP chromVAR scores are only calculated for ~200 TFs.
The text was updated successfully, but these errors were encountered: