formalization-pipeline

To convert qualitative disease descriptions into causal DAGs, this pipeline extracts causal claims, citations, and effect sizes from prose, then generates weighted DAGs with feedback-loop detection and TikZ/LaTeX figures. It also produces evidence inventories and validation reports with quantified uncertainty.

The best way to formalize ME/CFS pathophysiology prose is using a pipeline that transforms text into evidence-linked formal models. It produces DAGs, EPC event chains, and ODE-ready parameter artifacts with weighted certainty scores for direct manuscript integration.

Yes, you can generate ODE-ready parameter files from literature reviews. The pipeline prepares ODE parameter files when quantitative data permit, alongside EPC event chains with detection criteria and timings, yielding YAML model files ready for quantitative modeling.

Uncertainty quantification for extracted causal claims works by harvesting sample sizes and effect sizes from text, then assigning weighted certainty scores to each claim. The pipeline runs consistency and sensitivity checks to produce validation reports with quantified uncertainty.

The formalization pipeline supports both full disease reviews and single-process formalizations. It applies to subsystem analyses as well, producing evidence inventories, diagrams, and narrative summaries at the specified formalization level for document integration.

Automated EPC event chain construction depends on the availability of quantitative data in the source text to prepare ODE parameter files. While it generates event chains with detection criteria and timings, the certainty of causal claims is constrained by the extracted citations and sample sizes.