Please note: This PhD seminar will take place online.
Niousha Sadjadi, PhD candidate
David R. Cheriton School of Computer Science
Supervisor: Professor Lila Kari
Genomic signatures — taxon-specific nucleotide composition patterns — are used for taxonomic assignment and comparative genomics, yet their genome-wide pervasiveness across Telomere-to-Telomere assemblies remains underexplored. We address this gap with an alignment-free analysis using Frequency Chaos Game Representations (FCGRs) across the human genome and three other eukaryotes from distinct kingdoms. First, by combining qualitative inspection of FCGR landscapes with quantitative distance benchmarking, we show that each species exhibits a stable genomic signature across most chromosomes, with localized departures concentrated in tandem-repeat-rich regions. Then, we introduce two computational pipelines that select a short, contiguous representative genomic segment (500 Kbp) per genome that captures the genomic signature and can help quantify intragenomic variability and improve the performance of downstream applications such as one-nearest-neighbor taxonomic classification. Finally, we provide CGR-Diff, a graphical tool that enables visualization and quantitative comparison of FCGR-based genomic signatures. Overall, our results show that genomic signatures are pervasive yet predictably variable in repetitive regions and provide practical methods for representative segment selection and comparative analysis.