Hosted slide decks on Home-Cage Monitoring, FAIR metadata, virtual control groups, and preclinical data infrastructure — each a versioned, citable artifact with source provenance.
Each presentation is a versioned package: a structured deck, source map, speaker notes, and optional interactive visualizations. Built from the Neuronautix knowledge base.
A 15-slide ethics deck on the fourth obligation beyond the 3Rs: care for the data. From the moral contract of animal use to publication bias as a welfare problem, the ~$28B reproducibility cost, born-FAIR metadata as the memory of the experiment, shadow data, and the prerequisite for virtual control groups — closing on a new default: no animal-derived data without context.
A rigorous, metric-by-metric guide to Tecniplast DVC® analytics. From one raw signal — capacitance change across an under-cage electrode grid — it derives and cites each metric: Activation Density and Distance Walked, the activity budget and rest/activity bouts, the Rest Disturbance Index (RDI), circadian parameters with per-cage LED, spatial preference/frontality, the Bedding Status Index, the Urination Index, and GYM500 wheel running — closing on dark-phase signal, environmental channels as biomarkers, and data repurposing for the 3Rs.
COST TEATIME webinar deck arguing that FAIR data is the substrate that makes the 3Rs operational and the only credible road to AI-assisted research and Virtual Control Groups. Includes live demos of FAIR3R.fr and Metadatapp. Replay on the COST TEATIME YouTube channel; based on the WellFAIR paper (doi.org/10.1016/j.nsa.2026.106998).
Two root-level Markdown files that make any repository's FAIR posture and epistemic confidence readable in five seconds — for humans and machines alike. Covers the five claim categories, the 0–100 confidence scale, lineage from llms.txt to ClaimReview, and real corpus stats (19 notes · 440 claims · avg trust 74).
FAIR metadata management is part of animal welfare. This deck connects PREPARE, ARRIVE 2.0, and the 3Rs to a single data infrastructure need — and traces the path from born-FAIR workflows to virtual control groups.
An 18-slide vision of federated, interoperable, FAIR-by-default Home-Cage Monitoring — unified data streams, shared ontologies, open classifier infrastructure, and the 3Rs made structural — deliberately paired with a six-point critical sense-check of the limits that persist.
Invited SCAND-LAS 2026 talk arguing that FAIR metadata should be treated as a New Approach Methodology in its own right — the substrate that turns isolated preclinical data into reusable evidence and credible virtual control groups.
18-slide sequel to the May 18 KG-architecture deck — distinguishing graph DB / KG / ontology / GraphRAG, separating evidence-ontology-application layers, calibrating GraphRAG against vanilla RAG, treating link prediction as ranked hypothesis, and running four experiments before any custom model.
33-slide visual argument tracing the full path from the library-on-the-floor analogy through FAIR principles, WellFAIR, the 3Rs, and virtual control groups — with DVC® and Metadatapp as implementation anchors.
Long context is not understanding. A 20-slide visual argument for replacing universal-LLM-reader workflows with metadata, ontologies, graphs, validation, and targeted human review.
FAIR metadata is the enabling infrastructure that turns isolated preclinical data into reusable, AI-actionable evidence for animal reduction and scientifically credible virtual control groups.
A reworked, source-backed deck on turning preclinical metadata standardization into reusable, validated, and cautiously scoped virtual control workflows.
FAIR metadata enables virtual control groups that transform historical control data from a byproduct of each study into a shared, queryable asset — reducing the number of animals needed in future experiments.
Each deck is generated from the Neuronautix knowledge base, reviewed, and deployed as a versioned artifact. Source provenance, speaker notes, and claim citations travel with every slide.
Step 1Knowledge base retrieval — relevant notes and sources identified
Step 2Presentation brief — audience, duration, core message defined
Step 3Slide outline and spec reviewed before generation
Step 4Deck generated with inline citations and speaker notes
Step 5Human review via pull request before publication
Step 6Hosted at /presentations/slug/ with full source package
Work with Neuronautix
Request a presentation for your event
Neuronautix can prepare a tailored presentation for webinars, conferences, lab meetings, or training sessions. Topics include HCM system selection, FAIR metadata workflows, virtual control groups, and preclinical data infrastructure.
