GRAFIMO

GRAFIMO extends Position Weight Matrix (PWM) scanning to pangenome variation graphs (VGs) to identify transcription factor binding sites (TFBS) across genomes containing single nucleotide polymorphisms (SNPs), insertions and deletions (indels), and other variants.

Key Features:

• PWM scanning on graphs: Scans known transcription factor DNA motifs represented as Position Weight Matrices (PWMs) within pangenome variation graphs (VGs).
• Variant-aware analysis: Detects motif occurrences that are enhanced, weakened, or missed relative to the reference by accounting for SNPs, indels, and other variants.
• Pangenome encoding: Encodes collections of genomes and their variants within a compact VG data structure for population-scale representation.
• Multi-genome run: Performs a single run across multiple genomes embedded in the VG to identify motif occurrences across haplotypes.
• Haplotype and variant reporting: Reports motif frequency across different haplotypes and annotates occurrences that contain genomic variants or match the reference sequence.
• Statistical candidate selection: Produces detailed reports on statistically significant motif candidates.

Scientific Applications:

• Population TFBS analysis: Analyze transcription factor binding site variation across populations represented in pangenome VGs.
• Impact of genetic variation: Explore how SNPs, indels, and other variants affect TF binding landscapes and motif occurrence strength.
• Individual-specific regulation: Identify individual- or haplotype-specific TFBS that may contribute to phenotypic diversity or disease mechanisms.
• Large-scale genomic projects: Apply TFBS discovery across datasets from projects such as the 1000 Genomes project.

Methodology:

Scans PWMs across pangenome variation graphs (VGs) in a single run over multiple genomes/haplotypes and reports statistically significant motif occurrences with per-haplotype frequency and variant-versus-reference annotations.