Skip to main navigation Skip to search Skip to main content

Advancing genomic research in life sciences with highly accurate long-read DNA sequencing

Facility/equipment: Equipment › no e-resource/single sited

Equipments Details

Description

Recent advancements in DNA sequencing have had a profound impact on life science research, spanning microbial, fungal, plant, animal, and human genetics and genomics. Long-read, singlemolecule sequencing has emerged as a transformative technology, offering unprecedented capabilities for genome reconstruction and revealing intricate genome structures. The recently released PacBio VEGA DNA sequencing system represents an ideal platform to harness the power of highly accurate, long-read sequencing, while it is complementary in its throughput to sequencing platforms currently available in Flanders. The highly accurate long-read DNA of up to 20 kilobases allow reconstructing genomes, detecting large structural variations at an unprecedented level, giving invaluable insights into genome structure and evolutionary insights. It also enables studying organisms across the entire Tree of Life with larger genome sizes, enabling addressing research questions within a phylogenomic context. Moreover, its long reads facilitate high-resolution, amplicon-based taxonomic analysis of microbial ecosystems, offering valuable insights into intraspecies diversity and microbial ecology. Acquiring the PacBio VEGA will ensure that scientists at the forefront of their fields can continue performing cutting-edge and impactful research and accelerating scientific breakthroughs in a highly competitive (inter)national context.

Details

NameHet bevorderen van genomisch onderzoek in de levenswetenschappen met hoog-accurate en lange-fragment DNA-sequencing
Acquisition date1/05/26
Decommission date30/04/30

Keywords

  • Highly accurate long-read DNA sequencing
  • genome variation
  • phylogenomics

Fingerprint

Explore the research areas in which this equipment has been used. These labels are generated based on the related outputs. Together they form a unique fingerprint.