Past, ongoing, and future research plans

Some not-up-to-date lists of ongoing work to keep things organized for myself.

Dormant Link to heading

  • Minimizers:
    • publish the anti-lex scheme (→ WABI 26?)
    • continue investigation of greedymini to develop ‘understandable’ equivalent schemes.
    • Find exact optimal schemes for all \(k\equiv 1\pmod w\)
    • lower bound for local schemes

Long term plans Link to heading

If anything here interests you, feel free to reach out for collaborations.

  • engineer STPD (Becker et al. 2025): write a highly optimized implementation.
    • minimizer space?
    • fuzzy version by only using minimizer start positions?
  • Optimal CPU/memory performance given 3D latency and cooling constraints.
  • Pairwise alignment review paper based on thesis chapter 2 (this post).
    • Review on approximate string matching, this post (will never happen).
  • Minimizers review paper based on thesis part 2 (this post).

Abandoned Link to heading

  • Affine A*PA2: too much of an overhaul.
  • Spaced $k$-mer similarity: see this post.
  • Expected linear time A*PA: see a draft in this post.
  • Linear memory WFA: Instead of storing furthest reaching points for all wavefronts, it is sufficient to only store critical points where paths split/merge. This should lower memory usage of WFA to close to linear, without needing BiWFA. See this post. This has similar vibes to TALCO (Walia et al. 2024).
  • Local doubling: see this post.

References Link to heading

Becker, Ruben, Davide Cenzato, Travis Gagie, Ragnar Groot Koerkamp, Sung-Hwan Kim, Giovanni Manzini, and Nicola Prezza. 2025. “Compressing Suffix Trees by Path Decompositions.” arXiv; arXiv. https://doi.org/10.48550/ARXIV.2506.14734.
Walia, Sumit, Cheng Ye, Arkid Bera, Dhruvi Lodhavia, and Yatish Turakhia. 2024. “TALCO: Tiling Genome Sequence Alignment Using Convergence of Traceback Pointers.” In 2024 Ieee International Symposium on High-Performance Computer Architecture (Hpca). IEEE. https://doi.org/10.1109/hpca57654.2024.00044.