Skip to content
@unum-science

Unum Science

Fast tools & open datasets for the next wave of research: GPU sequence alignment, FASTA processing & billion-scale atlases. Built on @unum-cloud
README.md

Unum Science: fast tools and open datasets for the next wave of research

Unum on GitHub · Blog · Discord · YouTube · X · LinkedIn

Unum Science adapts the search, string & numeric engines of Unum to scientific & scholarly computing. Its tools run the classical algorithms of each field exactly, on GPUs and every CPU core, and its atlases publish what they compute in the open.

  • AffineGaps: exact sequence alignment & RNA folding with traceback on GPUs.
  • FasterFASTA: FASTA & FASTQ parsing, sorting & deduplication on every core.
  • USearchMolecules: open atlas of 7B+ molecules, proteins & antibodies.
  • USearchWiki: open atlas of wiki & scholarly text with embeddings & indexes.

Pinned Loading

  1. USearchMolecules USearchMolecules Public

    Large Multi-Modal Chem-Informatics dataset of 7B+ molecules with fingerprints and 3D conformers for drug discovery

    Python 112 11

  2. AffineGaps AffineGaps Public

    Exact biosequence alignment and folding on GPUs — Needleman-Wunsch, Smith-Waterman, Levenshtein, Zuker, and Sankoff with Gotoh adjustments and traceback

    Mojo 35 1

  3. FasterFASTA FasterFASTA Public

    Faster FASTA and FASTQ file processing command line tool - parsing, sorting, deduplication, transliteration, filter, and stats

    Rust 32 2

  4. USearchWiki USearchWiki Public

    Heavily enriched Wikipedia dataset, extending the HuggingFace's FineWiki with late-interaction (ColBERT-style) and auto-regressive (LLM-style) embeddings and additional metadata

    Python

Repositories

Showing 6 of 6 repositories

Top languages

Loading…

Most used topics

Loading…