1

Define the question

2

Assess fit and scope

3

Run literature and computational analysis where relevant

4

Deliver a written brief or dossier

Scientific research with molecular analysis equipment

Methods we may use

Not every project requires every method. The method depends on the question. The deliverable is always a written scientific asset.

  • Structured literature synthesis
  • Mechanism and pathway framing
  • Molecular interaction analysis where relevant
  • Network analysis
  • ADMET-style screening where useful
  • Structured evidence comparison

Methods validated in peer-reviewed venues

The computational approaches we may use have been validated through peer-reviewed publications at AAAI, ICLR, NeurIPS, and other top AI and bio venues.

AAAI-26 ORAL

Agentic Causal Graph Learning for Drug Target Discovery

David Scott Lewis, Enrique Zueco

Mechanism Mapping Causal Learning

Causal graph methods for mapping mechanism relationships — applied to nutrition pathway analysis.

Generative Models for Neuroprotective Compound Design: ADMET-Constrained Optimization

David Scott Lewis, Enrique Zueco

ADMET Screening Compound Safety

ADMET screening methods adapted for ingredient safety and bioavailability assessment.

NeurIPS AI4D3 POSTER

Active Causal Hypothesis Testing for AI-Guided Drug Target Discovery

David Scott Lewis, Enrique Zueco

Bioactive Analysis Hypothesis Testing

Validated causal hypothesis testing methods applied to bioactive compound analysis.

Multi-Scale Toxicity Screening via Verified Agentic Pipelines

David Scott Lewis, Enrique Zueco

Toxicology Safety Screening

Multi-scale safety screening of compounds — the same approach for ingredient-level analysis.

Cross-validation heatmap showing r>0.99 accuracy

Tell us the ingredient, fermentation, bioactive, or mechanism question you want analyzed.

Request a Brief