{
  "id": 18428596,
  "title": "Language-guided domain adaptation improves protein sequence rep-resentations for antimicrobial resistance analysis",
  "url": "https://urgent.news/2026/11/05/language-guided-domain-adaptation-improves-protein-sequence-rep",
  "topic": "science",
  "section": "Science",
  "published": "2026-11-05T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.11.04.764426v1?rss=1"
  },
  "original_language": "en",
  "account": "Antimicrobial resistance (AMR), especially antibiotic resistance in bacteria, poses challenges in combating infectious diseases. Protein sequences, responsible for antibiotic inactivation and efflux, offer a molecular basis for tracking resistance and guiding therapeutic interventions. Researchers have created BacLLM, a language-guided protein representation framework that incorporates functional knowledge derived from annotations into protein language model representations for AMR analysis in WHO priority bacterial pathogens.\n\nBacLLM enhances protein sequences, enriches UniProt annotations with semantic information, and employs cross-modal contrastive learning to integrate LLM-encoded annotation priors into the sequence encoder. In comparison to existing methods, BacLLM offers superior representations for identifying resistance genes and predicting mechanisms, detecting resistance-associated mutations at the single-residue level, and maintaining taxonomic relationships that are not evident through traditional clinical grouping methods based on phenotype.\n\nFurthermore, BacLLM successfully recovers resistance-related functional organization within a protein association network centered around the TolC protein, prioritizing proteins with low sequence identity to TolC but predicted structural similarity to it. This framework not only enhances AMR characterization but also aids in prioritizing therapeutic candidates for phage therapy. By improving strain-level phage host matching, BacLLM demonstrates its potential as a valuable tool for supporting AMR surveillance, functional characterization, and prioritizing therapeutic candidates.",
  "summary": "Antimicrobial resistance (AMR), particularly antibiotic resistance in bacterial pathogens, complicates infectious disease management. Because bacterial resistance is often mediated by protein functions, including antibiotic inactivation and efflux, protein sequence analysis provides a molecular basis for digital AMR surveillance and therapeutic intervention. We developed BacLLM, a language guided…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}