Genome-wide reconstruction of complex structural variants using read clouds

Abstract

Microfluidic partitioning of long genomic DNA fragments, and barcoding of shorter fragments derived from them, retains long-range information in short sequencing reads. Such read cloud approaches represent a powerful and cost-effective alternative to single-molecule long-read sequencing. We developed GROC-SVs, which uses read clouds for structural variant detection and assembly, and apply it to Illumina-sequenced 10× Genomics sarcoma and breast cancer data sets. Validation demonstrates substantial improvement in specificity of breakpoint detection compared to short-fragment sequencing, at comparable sensitivity, and vice versa. The long-range information also facilitates sequence assembly of breakpoints; importantly, consecutive breakpoints closer than the average length of the input DNA molecules can be assembled, with some events exhibiting remarkable complexity. We show that chromothriptic rearrangements occurred before copy number amplifications and that single-nucleotide and structural variants are not correlated. We predict significant advances in structural variant science using 10×/GROC-SVs and other read cloud-specific methods.

Comparative Annotation Toolkit (CAT) - simultaneous clade and personal genome annotation

“…However, this result is better than one that incorrectly represents any of the input haplotypes. Approaches that generate synthetic long reads, or read set technologies like 10X Genomics Linked Reads [84,85] combined with read-cloud alignment approaches [86], allow for approaches like Pilon to improve base level accuracy on regions not directly mappable by short read alignments. 2008 [39,63] A all-in-one pipeline that runs programs including AUGUSTUS and GeneWise with extrinsic information such as RNA-seq or protein sequences to both predict annotations and construct a gene set.…”

Haplotype-resolved and integrated genome analysis of the cancer cell line HepG2

“…The 10X Genomics linked-reads dataset was used to phase haplotypes as well as to identify, assemble, and phase primarily large (>30kb) SVs that include translocations (Spies et al 2017;Zheng et al 2016;Marks et al 2018). The 3kb-mate pair was used to validate the large SVs resolved from using the linked-read sequencing data and also to identify additional SVs, mostly in the medium size-range (1kb-100kb).…”

“…SVs and large-scale complex rearrangements were identified using both the Long Ranger wgs module with the "--somatic" option, GROC-SVs (default settings with breakpoint assembly) (Spies et al 2017), and gemtools (Greer et al 2017). The "--somatic" option increases the sensitivity of the large-scale SV caller for somatic SVs by allowing the detection of sub-haplotype events and does not affect small-scale variant calling.…”

Minerva: An Alignment and Reference Free Approach to Deconvolve Linked-Reads for Metagenomics

“…The core intuition behind this approach was that reads from the same fragment would tend to overlap with similar sets of reads that had different barcodes. Though each set of reads with the same barcode can contain reads from several fragments (5)(6)(7)(8)(9)(10)(11)(12)(13)(14)(15)(16)(17)(18)(19)(20), it is unlikely that two sets of reads with two different barcodes would contain reads from multiple separate overlapping fragments. If multiple reads with the same barcode overlapped with reads that had another barcode it would suggest that the original reads came from the same fragment.…”