Guiglielmoni, N. | Search bioRxiv

bioRxiv2020

Chromosight: A computer vision program for pattern detection in chromosome contact maps

Chromosomes of all species studied so far display a variety of higher order organizational features such as domains or loops often associated to biological functions and visible on Hi-C contact maps. We developed Chromosight, an algorithm inspired from computer vision that can detect patterns in Hi-C maps. Chromosight has greater sensitivity than existing methods, while being faster and applicable to any type of genomes, including bacteria, viruses, yeasts and mammals. Code and documentation: https://github.com/koszullab/chromosight

bioinformatics↗

bioRxiv2019

Chromosome-level quality scaffolding of brown algal genomes using InstaGRAAL, a proximity ligation-based scaffolder

Hi-C has become a popular technique in recent genome assembly projects. Hi-C exploits contact frequencies between pairs of loci to bridge and order contigs in draft genomes, resulting in chromosome-level assemblies. However, application of this approach is currently hampered by a lack of robust programs that are capable of effectively treating this type of data, particularly open source programs. We developed instaGRAAL, a complete overhaul of the GRAAL program, which has adapted the latter to allow efficient assembly of large genomes. Both GRAAL, and instaGRAAL use a Markov Chain Monte Carlo algorithm to perform Hi-C scaffolding, but instaGRAAL features a number of improvements including a modular polishing approach that optionally integrates independent data. To validate the program, we used it to generate chromosome-level assemblies for two brown algae, Desmarestia herbacea and the model Ectocarpus sp., and quantified improvements compared to the initial draft for the latter. Overall, instaGRAAL is a program able to generate, using default parameters with minimal human intervention, near-complete assemblies.

genomics↗