Transcriptional Core Regulatory Circuitry (CRC) analysis. CRC analysis leverages the observation that almost all known master transcription factors are regulated by large enhancers or super enhancers and that the master transcription factors bind to and regulate the enhancers of other master transcription factors. CRC analysis takes in a set of enhancer or other cis-regulatory regions and a list of active genes. For known transcription factors, CRC identifies those with proximal enhancers and learns how transcription factors are connected to one another. For each transcription factor, it does this by calculating the IN degree (the number of distinct transcription factors that regulate it) and the OUT degree (the number of other transcription factors regulated by the transcription factor in question). CRC analysis also determines which enhancers are bound by each transcription factor, and this can be useful to identify transcription factor cis-regulomes and gene expression programs.

The CRC software uses the following dependencies:

• Bamliquidator
• Samtools
• FIMO
• fasta-get-markov

Both FIMO and fasta-get-markov can be found in The MEME Suite.

pip install git+https://github.com/linlabcode/CRC.git

As a command line tool:

crc -e <enhacer_text_file> -g <genome_build> -s <subpeak_bed_file> -c <chromosomes_dir> -o <out_dir> -n <name>

As a python library:

import crc

crc.crc(enhancers, genome_input, chrom_path, output, analysis_name, bam=None, subpeak_file=None,
        mask_file=None, activity_path=None, const_extension=100, number=1, motifs=None,
        config='')

All example files were generated with data from a publicly available sample under the GEO accession number GSM2712746. Reads were aligned using the hg19 genome build.

REQUIRED ARGUMENTS

-e/--enhancer_file

ROSE2 generated enhancers table (_AllEnhancers_ENHANCER_TO_GENE.txt). Example

-g/--genome

Build of the genome to be used for the analysis. Currently supports HG38, HG19, MM10, and RN6.

-c/--chrom-path

Path to a folder with a seperate fasta file for each chromosome.

-o/--output

Output folder.

-n/--name

Name of the analysis.

ONE OF THE FOLLOWING ARGUMENTS IS REQUIRED

-b/--bam

Comma separated list of BAMs for valley finding.

-s/--subpeaks

BED file of regions to search for motifs. Example

OPTIONAL ARGUMENTS

-m/--mask

Masking file in BED format.

-a/--activity

Text file with active gene names. Either a list where each gene is in a new line or a table where one column holds gene names. Example

-l/--extension-length

Subpeak regions used in motif finding get extended by this length. Default is 100.

-N/--number

Number of non overlapping motifs in a region required to assign a binding event. Default is 1.

--motifs

An additional PWM file for the analysis.

--config

Genome configuration file to overwrite default paths.

_ADJ_LIST.txt

Adjacency list - the first label in a line is the source node and further labels in the line are target nodes.

_all_motifs.bed

BED file containing motif regions.

_all_subpeak.bed

BED file of all the subpeak regions.

_all_subpeak.fasta

FASTA file of all the subpeak regions.

_all_valleys.bed

BED file containing all valley regions used in motif search.

_bg.meme

1st order markov background file used in FIMO for motif search.

_CLIQUE_SCORES_DEGREE.txt

List of top 100 cliques or CRCs.

_DEGREE_TABLE.txt

File describing how many TFs a specific TF interacts with (out degree) and how many TFs interact with a specific TF (in degree).

_EDGE_LIST.txt

List of edges describing which TFs interact with each other.

_EDGE_TABLE.txt

Similar to '_EDGE_LIST.txt' but it also contains the interaction region and enhancer name.

_ENHANCER_TABLE.txt

Table of enhancers.

_ENHANCER_TF_TABLE.txt

Table connecting enhancers to TFs.

_ENRICHED_CLIQUE_FACTORS.txt

Node scoring; percentage of cliques the node is found in.

_GENE_SUMMARY.txt

Table describing if a gene is a TF and which enhancers are/could be connected to it.

_GENE_TABLE.txt

Table connecting genes and enhancers.

_GENE_TF_TABLE.txt

Table connecting TFs and enhancers.

_NODELIST.txt

List of nodes, the same as in '_ENRICHED_CLIQUE_FACTORS.txt'. These TFs are likely a part of a CRC.

FIMO

Folder containing FIMO command (bash script) and output.

Charles Y. Lin Baylor College of Medicine

Jost V. Koren Baylor College of Medicine

Alexander J. Federation University of Washington

Donald R. Polaski