genespots

genespots is an R script for identifying genomic hotspots—regions significantly enriched in genes of interest (GOIs)—using gene annotations from a GTF file and a provided list of GOIs (e.g., differentially expressed genes or gene clusters).

Workflow overview

• Maps GOIs to their genomic coordinates using the provided GTF file.
• Scans each chromosome using a sliding window (default: 250 kb) to estimate gene density.
• Identifies peaks in gene density as candidate hotspots.
• Assesses statistical significance by generating a null distribution via permutation of GOI labels (default: 100 permutations).
• Computes empirical p-values and adjusts for multiple testing using FDR (default threshold: < 0.05).
• Outputs:
  • A table of significant hotspots.
  • A PDF plot showing observed densities, null densities, and hotspot locations.

Installation

You can run this script locally in R or RStudio. To clone the repository:

git clone https://github.com/ccarloscr/genespots.git

Sample Inputs

The Input folder contains example files:

• A GTF annotation file dmel-all-r6.62.gtf
• A table of GOIs defs_degs.txt Make sure the GTF file is decompressed before use.

Configuration

You can customize parameters by editing the script directly:

• gtfdir: path to the GTF annotation file.
• clusterdir: path to the table of genes of interest.
• bandwidth: size of the sliding window.
• n_perm: number of permutations for null distribution.
• min_density_threshold: fraction of max density used to filter hotspot peaks.
• min_genes: minimum number GOIs overlapping a sliding window to consider a valid hotspot.
• filter_chromosomes: logical flag to restrict plots to chromosomes with sig. hotspots.

License

This project is licensed under the MIT License.