[Building Optimized Units of Quantified Enhancer Topologies]
Explore the docs »
Read the paper »
Request Feature
|
Report Bug
⭐ Consider starring the repo! ⭐
SOFTWARE AUTHORS: Jie Lv, Brian J Abraham
jie.lu@stjude.org
brian.abraham@stjude.org
Installing BOUQUET:
git clone https://github.com/stjude/BOUQUET.git; cd BOUQUET
BOUQUET is an integrative, graph-theory-based approach that uses multiple aspects of genome topology to probe communities of CREs, their bound apparatus, and their target genes. It is a community-detection algorithm that uses concepts from graph theory to convert undirected networks of CREs and their connections (“nodes” and “edges”, respectively) into discrete communities of CREs associated with each expressed gene.
Briefly, it is based on a modified label-propagation strategy wherein highly interconnected sets of nodes are unbiasedly assembled into communities. Any additional CREs connected directly to a given promoter either by high-confidence HiChIP loops or shared INs are then incorporated into that gene’s community. Finally, any leftover CREs are added to a gene's community by their concurrence within the same TAD.
usage : bash 3DSE.wraper.sh -o OutDir [-d ScriptDir] [-p Promoters] [-I INs] [-l Loops] [-e Enhancers] [-s EnSignal] [-c EnControl] [-g HighlightGenes] [-h]
Use option -h for more information
Options:
-o OutDir user specified directory for all output of the pipeline, recommend to reflect the unique combination of input files
-d ScriptDir directory for all 3DSE scripts
-p Promoters list of gene promoters, in the format of Chr, Start, End, TranscriptID|GeneID, default to be 4Kb region around TSS
-w WholeGenes list of whole gene body, in the format of Chr, Start, End, TranscriptID|GeneID, default to be whole gene body
-I INs Isolated Neighborhoods, in the format of Chr, Start, End.
-l Loops Direct loops connecting two regulatory regions, in the format of bedpe. Default to be the ones defined from H3K27ac HiChIP.
-e Enhancers Set of enhancers in the format of Chr, Start, End. Default to be the ones defined from H3K27ac ChIP-Seq.
-s EnSignal ChiP-seq signal to quantify Enhancer signal intensity in the number of reads. The default is to be the Bam file of MED1 ChIP sample.
-c EnControl Control for ChiP-seq signal to quantify Enhancer signal intensity in the number of reads. Default to be Bam file of MED1 Control(Input) sample.
-g HighlightGenes a list of genes for which comprehensive GRN figures and tracks are to be generated. Expecting format 'GeneSymbol1|RefSeqID1,GeneSymbol2|RefSeqID2...'. Please note a single quote is needed because of the special character "|".
-n Network a switch indicating whether to draw network figures (7) for highlighted genes. could be slow for genes with complex landscapes. Default to turn off.
-O OtherEnd a switch indicating whether to perform functional characterization for promoter-interacting regions.
-f LO a switch indicating whether to use both loops and INs-supported regulatory edges (default, 0) or only loop-supported regulatory edges (1).
-F PromIN a switch indicating whether to only assign E to P (default, 0) or assign both E and P to P (1) according to INs (default, 0).
-v H3K27acPeaks H3K27ac peaks to define enhancers
-x H3K4me3Peaks H3K4me3 peaks to define promoters
-y CTCFPeaks CTCF peaks to define insulators
-z H3K27me3Peaks H3K27me3 peaks to define PcG sites
-E ExpFile gene symbol-level expression file calculated based on RNA-Seq
-S Seed Seed number of R random process. same seed value would generate reproducible gene communities based on the LP method.
-t TAD Topological Associated Domain used for rescue EO (Enhancer Orphan) after community-based Enhancer assignment