LINKS 1.6.1 (Feb 24, 2016)

Incorporation of the BC Genome Sciences Centre custom Bloom filter with the recursive nucleotide hash function (ntHash).

For additional information about this project, please visit the overview page .

Available downloads

links_v1-6-1.tar.gz

For all platforms (1.0 MB)

Release Notes

Incorporation of the BC Genome Sciences Centre custom Bloom filter with the fast ntHash recursive nucleotide hash function. This new data structure supports the creation of canonical Bloom filters from k-mers of large genome assemblies. We tested building Bloom filters from the 3 Gbp human and 20 Gbp white spruce genome assemblies and clocked the processes to 39m (26GB RAM) and 5h (178GB RAM), respectively. The Bloom filter data structure swap in v1.6 offers a 30-fold kmer insert speed-up (~6x query speed-up) over v1.5.2, while supporting the creation of filters from very large genome assembly drafts. Additional, human-readable correspondence output file (.assembly_correspondence.tsv) listing the scaffold and contig ID along with original contig assembly names, number of linking k-mer pairs and estimated gap lengths. Description on how to interpret the graph / .gv file (see LINKS-readme.txt; Interpreting the graph / .gv file)