Login / Signup

Trycycler: consensus long-read assemblies for bacterial genomes.

Ryan R WickLouise M JuddLouise T CerdeiraJane HawkeyGuillaume MéricBen VezinaKelly L WyresKathryn E Holt
Published in: Genome biology (2021)
While long-read sequencing allows for the complete assembly of bacterial genomes, long-read assemblies contain a variety of errors. Here, we present Trycycler, a tool which produces a consensus assembly from multiple input assemblies of the same genome. Benchmarking showed that Trycycler assemblies contained fewer errors than assemblies constructed with a single tool. Post-assembly polishing further reduced errors and Trycycler+polishing assemblies were the most accurate genomes in our study. As Trycycler requires manual intervention, its output is not deterministic. However, we demonstrated that multiple users converge on similar assemblies that are consistently more accurate than those produced by automated assembly tools.
Keyphrases
  • single molecule
  • patient safety
  • high resolution
  • machine learning
  • deep learning
  • gene expression
  • clinical practice
  • high throughput
  • single cell
  • mass spectrometry
  • dna methylation
  • electronic health record