Adnan KhanAdnanProduct & AI
San Francisco Bay Area
Get in touch

Source note3 min read

LLM-Guided ODE Discovery and Parameter Inference from Small-Cohort Aggregate Data

Source

arXiv

arxiv.org/abs/2607.00733v1

Read the source

Source

Plain-Language Brief

A new method called AgentODE has been proposed to discover and refine Ordinary Differential Equations (ODE) structures for modeling complex dynamics, particularly useful in clinical settings. This method works with small-cohort aggregate data, which is often the case in rare diseases where individual-level data is scarce and privacy concerns are high. The method uses a Language Model (LLM) to propose candidate ODE structures and an iterative refinement process to refine parameter distributions using summary statistics alone. The method was evaluated on three benchmark problems and two clinical datasets, including recessive dystrophic epidermolysis bullosa (RDEB), with only 231 observations across 46 patients. The results showed that AgentODE can recover functionally consistent ODE structures across all settings and performs better in sparse and noisy data settings compared to baselines that have access to individual-level data.

What the Document Says

  • AgentODE is a framework for discovering and refining Ordinary Differential Equations (ODE) structures from population-level summary statistics.
  • The method uses an LLM to propose candidate ODE structures.
  • An iterative refinement process is used to refine parameter distributions through a diagnosis--update loop.
  • The method was evaluated on three benchmark problems and two clinical datasets, including RDEB with 231 observations across 46 patients.
  • AgentODE can recover functionally consistent ODE structures across all settings.

Key Concepts Mentioned

  • Ordinary Differential Equations (ODE)
  • Mechanistic modeling
  • Rare diseases
  • Summary statistics
  • LLM
  • Recessive dystrophic epidermolysis bullosa (RDEB)

Claims Made by the Document

  • AgentODE addresses the gap of jointly discovering ODE structure and refining parameter distributions solely from summary statistics.
  • Reasoning from summary statistics promotes mechanistically principled structure discovery in sparse and noisy data settings.

Evidence Used

  • The text states that AgentODE is designed to work with population-level summary statistics, capturing heterogeneity by modeling parameters as distributions rather than fixed values.
  • The text mentions an iterative refinement process using a diagnosis--update loop operating on population-level summary statistics.

Methods / Approach

UNKNOWN

Limitations Stated by the Authors

UNKNOWN

Limitations Noted by the Agent

Agent interpretation — not from the source.

UNKNOWN

Open Questions

UNKNOWN

Notes for Future Q&A

May include interpretation. Labeled where it does.

UNKNOWN

Raw Abstract / Excerpt

Mechanistic modeling via ordinary differential equations (ODEs) provides interpretable descriptions of complex dynamics and enables inference of underlying mechanisms, which is particularly valuable in clinical settings. However, in rare diseases, both the structure and parameters of the model are typically unknown, while individual-level data is scarce, noisy, heterogeneous, and subject to privacy constraints. In such settings, population-level summary statistics provide a practical privacy-preserving data representation, while capturing heterogeneity further requires modeling parameters as distributions rather than fixed values. Yet no existing method jointly discovers ODE structure and refines parameter distributions solely from summary statistics. We present AgentODE, an end-to-end framework that addresses this gap. An LLM proposes candidate ODE structures, while a tool-augmented inference agent iteratively refines parameter distributions through a diagnosis--update loop, operating on population-level summary statistics alone. We evaluate AgentODE on three benchmark problems across different fields and two clinical datasets, including the rare disease recessive dystrophic epidermolysis bullosa (RDEB), with only 231 observations across 46 patients. AgentODE recovers functionally consistent ODE structures across all settings, and experiments on RDEB demonstrates that in sparse and noisy data settings reasoning from summary statistics promotes mechanistically principled structure discovery, whereas baselines with individual-level data access recover implausible structures despite better predictive performance. AgentODE opens new possibilities for mechanistic modeling of rare diseases directly from population-level summary statistics, where data scarcity and privacy constraints have traditionally limited such analyses.