Science ▸ Methodology
How MESSAI turns literature into calibrated predictions
The one technical-reference page for the whole method — from the first principles of bioelectrochemistry (thermodynamics, electrode kinetics, how bacteria move electrons) through the data pipeline, the hierarchical Bayesian priors, and the calibration layer. Depth-layered: skim the intuition, or expand the full method. Honest about its limits — where a layer is a 0-D mean function or plain split-conformal, it says so.
Overview
This is the methodology page — the honest, end-to-end account of how MESSAI turns a corpus of literature into calibrated, uncertainty-aware predictions. It runs from first principles (thermodynamics, kinetics, biology) through the data pipeline, the Bayesian priors, and the calibration layer. Where a layer is a simple 0-D mean function or plain split-conformal calibration, it says so plainly. The honesty is the point.
A microbial electrochemical system (MES) — the umbrella also called a bioelectrochemical system (BES) — exploits the electrochemical activity of living microorganisms to interconvert chemical and electrical energy, or to drive chemical synthesis. The field is unified by a single idea: use microbes as self-replicating electrocatalysts. Every device in the taxonomy is a variation on one cell; what changes is the reaction you drive and the product you harvest.
In the canonical case — a microbial fuel cell — electroactive bacteria colonise an anode and oxidise an organic substrate such as acetate or the organic load of wastewater. Oxidation releases electrons, which the bacteria deposit onto the electrode instead of onto a soluble acceptor. Those electrons travel through an external circuit to a cathode, where they reduce a terminal acceptor (often oxygen). The electron flow is a usable current; the proton flow that balances it closes the circuit through the electrolyte.
The core mechanism
Bacteria at the anode oxidise substrate and release electrons. Electrons flow through the external circuit to the cathode, generating power while reducing an acceptor. Protons migrate through the electrolyte or a membrane to balance charge.
Why microbes
Microbes are self-replicating, self-repairing catalysts that work at ambient temperature and pressure and can process complex, dilute waste streams that precious-metal catalysts handle poorly or not at all.
Two facts shape everything downstream, and are worth stating up front because they explain *why* the method looks the way it does. First, the literature reports the same quantity in mutually incompatible ways — different normalisation bases, peak vs steady-state, undeclared reference electrodes — so raw numbers cannot simply be averaged. Second, the headline metrics are heavy-tailed across orders of magnitude, so a single number is almost always the wrong answer. The rest of this page is the disciplined response to those two facts: classify precisely, extract with provenance, harmonise ruthlessly, model on the log scale, and report calibrated distributions rather than point values.
Depth-layered by design: skim the leads, tables, and figures; expand the "Read the full method" panels for schemas, equations, and diagnostics. Numbers are dated snapshots from a shared source and link out to the live surfaces — the proof dashboard for provenance, insights for the corpus rollup.