Skip to content

Release Notes

All notable changes to SeroRef are documented on this page. This project adheres to Semantic Versioning and the Keep a Changelog specification.

[Unreleased]

Added

  • Single-Pass ORF & CNN Role Engines (seroref.genes):
  • 29-state frameshift- and operon-overlap-aware profile-HMM Viterbi OrfPredictor with sliding 4-kb GC adaptation and single-pass viral/prophage gap rescue (seroref.genes.orf.predictor).
  • 30-class GeneRole ontology (29 functional polysaccharide biosynthesis, transport, modification, regulatory, and MGE roles + open-set UNKNOWN) and pure NumPy/Numba multi-scale 1D-CNN GeneRoleClassifier with a joint 1-bit bacterial vs. bacteriophage/prophage origin head (seroref.genes.cnn).
  • De Novo Polysaccharide Locus Discovery (seroref.loci):
  • 11-class LocusArchetype and ArchetypeCategory ontology (seroref.loci.archetypes) and ArchetypeClassifier / LocusClassifier (seroref.loci.classifier).
  • Exact All-Contiguous-Partitions Dynamic Programming (ACP-DP) + multi-locus Viterbi LocusSegmenter (seroref.loci.segmenter).
  • LocusPredictor with TRANSPOSASE/INTEGRASE MGE insertion bridging, optional coordinate healing (heal_mobile_elements), and cross-contig boundary stitching (stitch_contig_splits).
  • LocusMatcher for de novo orthogroup database construction and MatchType structural variant bitflag classification (seroref.loci.matcher).
  • 4-Tier De Novo Reference Database (seroref.db):
  • SeroRefRecord and SeroRefRecords Structure-of-Arrays container (src/seroref/data/seroref.npz) linking organism, reference strain, DSMZ StrainInfo/BacDive equivalence cliques (strain_synonyms), NCBI assembly/nuccore accession, serotype/glycan antigen token, and de novo predicted Locus.
  • 4-tier async curation pipeline (seroref.db.curation: ENANCTC3000Tier, StrainInfoTier, TaxonCatalogTier, and de novo target resolution via seed_discovery_targets_from_sources, ExplicitAccessionTier, PubMedElinkTier, and StrainAssemblyTier).
  • Interactive Benchmarks & Documentation:
  • Reproducible 16:9 Marimo slide decks (curate_seroref.py, benchmark_prophage_orfs.py, train_gene_classifier.py, benchmark_spneumo_db.py) and unified SeroVerse Zensical documentation.

Comments