CrossStitch creates personalized reference-quality diploid genomes without de novo assembly. The basic idea is rather than trying to assemble a genome from scratch, it will leverage a reference genome as a baseline, and then update it with any SNPs, indels, or structural variations present in your sample. For the best results, the data requirements are similar to a de novo assembly: Illumina-based data for SNPs and Indels, Long Read data for structural variants, and Phasing data such as 10X Linked Reads and/or HiC data. However the CrossStitch process is much less demanding, produces more accurate results, and the process is much more predictable. The output will be a phased VCF file with all variants (SNPs, Indels, and SVs) as well as a phased personalized diploid genome including 2 copies of each chromosome with the variants inserted at the correct locations.

Software versions

Download checksum
commit 53f64af