Using the Model Context Protocol (MCP) open standard, mzassist connects mzmine to a lab’s approved MCP-compatible AI assistant.
Users can set up and run workflows in natural language, from importing raw data and configuring parameters to running batches and summarising results, improving consistency and freeing time for scientific work.
Raw mass spectrometry data stays on the lab’s computer and users control what information is shared with AI.
AI assistance is active only when enabled: users can delegate the workflow or intervene manually at any stage.
Deterministic mzmine algorithms — not the language model —make parameter decisions ... and labs connect an assistant approved by their IT and data protection teams.
“The lab chooses which AI assistant to trust, and mzmine stays the authoritative environment for the data analysis itself,” says Dr Ansgar Korf, CEO and founder of mzio.
“That combination makes AI practical in a mass spec workflow without compromising reproducibility or the vendor-neutral position that mzmine has held since 2005.”
mzassist helps users adopt existing and new analytical methods by explaining processing steps and parameters and supporting spectral library building.
Its parameter estimator derives settings from selected raw data files; the auto optimiser benchmarks them on the actual data.
Each analysis is recorded in a complete mzmine workflow file, preserving every processing step and parameter for reproducibility and traceability.
“Parameter estimation and optimisation in mzassist are data-driven algorithms, not guesswork by a language model,” says Dr Steffen Heuckeroth, CTO and founder of mzio.
“The estimator derives parameters directly from the signals in a lab’s raw data, with consistent results every time.”
mzassist is included in mzmine COMMUNITY for academic users and mzmine PRO for commercial customers.
Labs can choose an AI assistant that meets their IT, data protection and compliance requirements.
