Cooccurrence

The cooccurrence pipeline is designed to calculate pairwise cooccurrence of features within GFF files, facilitating the analysis of their relationships. The pipeline leverages the bedtools closest function to determine pairwise relations based on specific constraints, such as one feature occurring upstream of another. Subsequently, maximal cliques are determined, and a combinatorial table is generated containing all identified groups. This workflow is particularly useful for investigating feature interactions and cooccurrences in biological data.

Features

• Nextflow pipeline for calculating pairwise cooccurrence of features within GFF files.
• Utilizes bedtools closest to establish pairwise relationships, allowing detailed constraints for feature pairs.
• Maximal cliques are identified to group cooccurring features.
• Option to generate a CSV file containing all complete graphs (combinations).
• Visualization of the identified cooccurring feature groups is available.

How to Use

To calculate all complete graphs (combinations) and return them as a CSV:

nextflow main.nf

To additionally visualize the identified cooccurring feature groups:

nextflow main.nf -entry plot

Requirements

• General: nextflow version 21.04.3.5560 or later.

Notes

• Input files are expected to be in the GFF format and are specified within the script.
• Parameters for bedtools closest and maximal distances are provided through a combinatorial table in CSV format.
  • Default parameters can be set within the script.
  • Non-default lines should be configured in the provided configuration.
  • Use simplified file names (e.g., myfile.gff -> myfile) as identifiers for the combinatorial table.
• The script builds undirected edges for K2 graphs; however, symmetry is not validated.
• Overlapping intervals are not handled in detail.
• Example data is from RegulonDB v 10.5 (Alberto Santos-Zavaleta et al. 2019).