LPHash - datasets

These datasets accompany the paper "Locality-Preserving Minimal Perfect Hashing of k-mers", G. E. Pibiri, Y. Shibuya, and A. Limasset, 2023, Bioinformatics DOI: https://doi.org/10.1093/bioinformatics/btad219.

We use genomes of increasing size in terms of number of distinct k-mers; namely, the whole- genomes of:

Saccharomyces Cerevisiae (Yeast, 11.6 × 10^6 k-mers),

Caenorhabditis Elegans (Elegans, 95×10^6 k-mers),

Gadus Morhua (Cod, 0.56×10^9 k-mers),

Falco Tinnunculus (Kestrel, 1.16×10^9 k-mers), and

Homo Sapiens (Human, 2.77 × 10^9 k-mers).

For each genome, we obtain the corresponding SPSS (Spectrum-Preserving String Set) by first building the compacted de Bruijn graph using BCALM2 (Chikhi et al., 2016), then running the UST algorithm (Rahman et al., 2020). At our code repository (github.com/jermp/lphash) we provide detailed instructions on how to obtain SPSS datasets like the ones available here from fasta files.

References

Rayan Chikhi, Antoine Limasset, and Paul Medvedev. Compacting de Bruijn graphs from sequenc- ing data quickly and in low memory. Bioinformatics, 32(12):i201–i208, 2016.

Amatur Rahman and Paul Medvedev. Representation of k-mer sets using spectrum-preserving string sets. In International Conference on Research in Computational Molecular Biology, pages 152–168. Springer, 2020.