AccuSyn: Using Simulated Annealing to Declutter Genome Visualizations
Date
2019-09-24
Authors
Journal Title
Journal ISSN
Volume Title
Publisher
ORCID
Type
Thesis
Degree Level
Masters
Abstract
We apply Simulated Annealing, a well-known metaheuristic for obtaining near-optimal solutions to optimization problems, to discover conserved synteny relations (similar features) in genomes. The analysis of synteny gives biologists insights into the evolutionary history of species and the functional relationships between genes. However, as even simple organisms have huge numbers of genomic features, syntenic plots initially present an enormous clutter of connections, making the structure difficult to understand. We address this problem by using Simulated Annealing to minimize link crossings. Our interactive web-based synteny browser, AccuSyn, visualizes syntenic relations with circular plots of chromosomes and draws links between similar blocks of genes. It also brings together a huge amount of genomic data by integrating an adjacent view and additional tracks, to visualize the details of the blocks and accompanying genomic data, respectively. Our work shows multiple ways to manually declutter a synteny plot and then thoroughly explains how we integrated Simulated Annealing, along with human interventions as a human-in-the-loop approach, to achieve an accurate representation of conserved synteny relations for any genome. The goal of AccuSyn was to make a fairly complete tool combining ideas from four major areas: genetics, information visualization, heuristic search, and human-in-the-loop. Our results contribute to a better understanding of synteny plots and show the potential that decluttering algorithms have for syntenic analysis, adding more clues for evolutionary development. At this writing, AccuSyn is already actively used in the research being done at the University of Saskatchewan and has already produced a visualization of the recently-sequenced Wheat genome.
Description
Keywords
synteny, comparative-genomics, simulated-annealing, heuristic-search, javascript, d3-visualization, circos
Citation
Degree
Master of Science (M.Sc.)
Department
Computer Science
Program
Computer Science