[
    {
        "id": "osp-15894",
        "type": "article-journal",
        "title": "APOD: reasoning-guided agentic population ordinary differential equation discovery for pharmacological digital twins",
        "author": [
            {
                "family": "Ferrara",
                "given": "Romain"
            },
            {
                "family": "Soucail",
                "given": "Martin"
            },
            {
                "family": "Gertner",
                "given": "Victor"
            },
            {
                "family": "Moussali",
                "given": "Adil"
            },
            {
                "family": "Cocquebert",
                "given": "Joris"
            },
            {
                "family": "Oziel-Taieb",
                "given": "Sandrine"
            },
            {
                "family": "Nicolas",
                "given": "Julien"
            },
            {
                "family": "Gattacceca",
                "given": "Florence"
            },
            {
                "family": "Schaar",
                "given": "Mihaela van der"
            },
            {
                "family": "Benzekry",
                "given": "Sébastien"
            }
        ],
        "URL": "https://omanscience.com/ar/articles/apod-reasoning-guided-agentic-population-ordinary-differential-equation-discovery-for-pharmacological-digital-twins",
        "language": "en",
        "issued": {
            "date-parts": [
                [
                    2026
                ]
            ]
        },
        "abstract": "Establishing ordinary differential equations (ODEs) describing population data is a fundamental part of mathematical modeling in pharmacology, crucial to developing digital twins. However, doing so from sparse, noisy data is a slow, expert-driven task. Existing automated methods either search a restricted model space or ignore population inter-individual variability. Here we introduce APOD (Agentic Population ODE Discovery), a language-model agent that iteratively reasons over biological knowledge and fit diagnostics in an open-ended search space to discover a population digital twin (PDT), i.e., a shared ODE system with between-subject variability. On synthetic pharmacokinetic and tumor-dynamics benchmarks, APOD recovered ground-truth structures in 94-100\\% of runs, 12-fold faster in median than an established library-based search. On real cohorts it converged to valid structures, and proposed a PDT of radioligand-therapy-induced platelet dynamics that predicts thrombocytopenia from first-cycle data and simulates alternative dosing schedules that lower the predicted risk of toxicity."
    }
]