{
  "id": 6097777,
  "title": "From code to natural language: MErlin - a multiomics toolkit for bacterial epigenomics delivered as Claude agent skill.",
  "url": "https://urgent.news/2026/09/06/from-code-to-natural-language-merlin-a-multiomics-toolkit-for",
  "topic": "ai",
  "section": "AI",
  "published": "2026-09-06T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.09.02.748773v1?rss=1"
  },
  "original_language": "en",
  "account": "MErlin is a cutting-edge multi-omics toolkit designed to analyze bacterial methylomes. This powerful tool integrates various data types, including modified-base calls, genome annotations, motif inventories, methyltransferase genotypes, transcript abundance, replicon positions, and chromosome conformation. To address the challenges of incompatible formats, inconsistent identifiers, and diverse analytical workflows, MErlin comprises seventeen composable modules that seamlessly convert raw data into ranked gene-level evidence and a comprehensive HTML report.\n\nMErlin is available in two formats: as a standard Python package and as an agent skill. The agent skill is a unique, version-controlled layer of procedural knowledge that allows a compatible large language model (LLM) to seamlessly select and execute the necessary analytical operations without the need for developers to create a new analysis implementation for each request. This innovative design treats natural language as an interface to fixed analytical operations, rather than a substitute for proven scientific software.\n\nThe skill behind MErlin encodes module-selection rules, mandatory pre-flight checks, human input requirements, design-to-inference constraints, and interpretation guidance. This ensures that users receive accurate and reliable results while maintaining a user-friendly experience. MErlin's architecture and statistical methods were validated using a synthetic dataset with planted ground truth, guaranteeing its robustness and reliability.\n\nTo demonstrate the conversational interface, MErlin was applied to a real-world methylome-transcriptome comparison in Pseudoalteromonas haloplanktis TAC125. The toolkit's ability to streamline complex bacterial epigenomics analyses makes it an invaluable resource for researchers in the field, providing a powerful, user-friendly solution to an intricate problem. MErlin is open source and available at https://github.com/IacopoPasseri/MErlin.",
  "summary": "Interpreting a bacterial methylome is a multi-omics problem. It requires integrating modified-base calls with genome annotation, motif inventories, methyltransferase genotypes, transcript abundance, replichore position and, increasingly, chromosome conformation. These data types are commonly generated in incompatible formats, use inconsistent sequence and gene identifiers, and originate from…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 2,
    "also_reported_by": [
      {
        "outlet": "Dev.to",
        "title": "Keeping Claude Code sessions for your repository in one place",
        "url": "https://urgent.news/2026/09/06/keeping-claude-code-sessions-for-your-repository-in-one-place",
        "published": "2026-09-06T04:17:55.000Z"
      }
    ]
  },
  "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."
}