Skip to content

Latest commit

 

History

History
7 lines (5 loc) · 1.2 KB

README.md

File metadata and controls

7 lines (5 loc) · 1.2 KB
Raw

Supplementary material for the PDX Hi-C paper

Work in progress.

  • 00_preprocessing.Rmd - Separates loop data in any text format into PR/CR-specific and common loops and anchors. If merge_adjacent <- TRUE, nearby (flanking) regions are considered common. The code chunks in the "Load data" section create two GInteraction objects non-overlapping with excludable regions (centromeres, telomeres, etc.), which are processed using the same downstream code. Only one code chunk corresponding to a specific data type should be set to eval = TRUE. Data from Mustache, HiCcompare, NeoLoopFinder, GENOVA, SpectralTAD, and hicFindTADs can be loaded. Data loading chunks for other data formats can be written.
    • Input: Loop data. Set the corresponding code chunk to eval=TRUE
    • Output: Separated loops and anchors in the preprocessing subfolder. PR/CR all, common, and unique loops/anchors are outputted as BEDPE/BED files. Color-specifying columns are included to emphasize PR/CR/common loops/anchors (green/red/blue colors). Loops/anchors are not collapsed (reduced); therefore, PR_common and CR_common loops/anchors may differ. log.txt - counts and width of loops and anchors.