Obtaining highly accurate and contiguous sequence assemblies is a key goal for improved characterization of individual and reference genomes, particularly in uncovering complex rearrangements and amplifications, and in understanding the function of repetitive sequences and the non-coding genome. We introduce CROM, a super-scaffolding approach that directly utilizes raw single molecule optical maps to cost-effectively reconstruct arm-level assemblies of large eukaryotic genomes. CROM is based on the scaffold overlap-extension-bridge paradigm and combines a sensitive and precise map-to-sequence alignment with a local error minimizing consensus stage and a combinatorial approach to scaffold bridging. CROM's flexible framework also allows for the detection of alternative haplotypes. Benchmarking CROM against state-of-the-art methods showed that it provides significant improvements in contiguity (20–212%, NG50) while maintaining high-quality assemblies. We show that the application of CROM enables spanning of human genome hotspots, improved placement of previously unlocalized human reference scaffolds and the best de novo human assembly to date (43 Mb scaffold NG50 corresponding to 73% of GRCh38 NG50) using only two single-molecule clone-free technologies. CROM is the first method that can successfully exploit different genome-mapping technologies, allowing for the reuse of datasets for improving draft genomes.
Super-Scaffolding of Large Eukaryotic Genomes with Single Molecule Maps
VERZOTTO D;
2016-01-01
Abstract
Obtaining highly accurate and contiguous sequence assemblies is a key goal for improved characterization of individual and reference genomes, particularly in uncovering complex rearrangements and amplifications, and in understanding the function of repetitive sequences and the non-coding genome. We introduce CROM, a super-scaffolding approach that directly utilizes raw single molecule optical maps to cost-effectively reconstruct arm-level assemblies of large eukaryotic genomes. CROM is based on the scaffold overlap-extension-bridge paradigm and combines a sensitive and precise map-to-sequence alignment with a local error minimizing consensus stage and a combinatorial approach to scaffold bridging. CROM's flexible framework also allows for the detection of alternative haplotypes. Benchmarking CROM against state-of-the-art methods showed that it provides significant improvements in contiguity (20–212%, NG50) while maintaining high-quality assemblies. We show that the application of CROM enables spanning of human genome hotspots, improved placement of previously unlocalized human reference scaffolds and the best de novo human assembly to date (43 Mb scaffold NG50 corresponding to 73% of GRCh38 NG50) using only two single-molecule clone-free technologies. CROM is the first method that can successfully exploit different genome-mapping technologies, allowing for the reuse of datasets for improving draft genomes.| File | Dimensione | Formato | |
|---|---|---|---|
|
16-RECOMB_2016.pdf
non disponibili
Licenza:
NON PUBBLICO - Accesso privato/ristretto
Dimensione
1.34 MB
Formato
Adobe PDF
|
1.34 MB | Adobe PDF | Visualizza/Apri Richiedi una copia |
I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
