{
  "id": 15856554,
  "title": "Fitting and Selection of Within-Host Models for SARS-CoV-2 Infection",
  "url": "https://urgent.news/2026/10/20/fitting-and-selection-of-within-host-models-for-sars-cov-2-infection",
  "topic": "science",
  "section": "Science",
  "published": "2026-10-20T00:00:00.000Z",
  "source": {
    "name": "bioRxiv",
    "slug": "biorxiv",
    "url": "https://www.biorxiv.org/content/10.64898/2026.10.15.759195v1?rss=1"
  },
  "original_language": "en",
  "account": "Selecting adequate mathematical models to describe within-host SARS-CoV-2 infection is crucial when analyzing longitudinal RT-qPCR data. Researchers evaluated three within-host models - target-cell-limited (TIV), eclipse-phase extension (TEIV), and fractional-order TIV - by fitting them to SARS-CoV-2 cycle-threshold (Ct) measurements from a National Basketball Association cohort. The models were optimized by maximizing a right-censored Gaussian likelihood function using hybrid global-local optimization methods.\n\nModel selection relied primarily on the small-sample corrected Akaike information criterion (AICc), derived from the censored likelihood. The researchers also examined the robustness of their choices using the Akaike information criterion (AIC) and Bayesian information criterion (BIC). In 34.0% of the 868 evaluable trajectories, the unpenalized likelihood analysis identified TEIV as the optimal model, while in 30.2% of cases, fr-TIV emerged as the best-fitting model, and in 3.3%, TIV was the preferred choice. In the remaining 32.5% of trajectories, the models were effectively tied.\n\nUpon accounting for model complexity, AICc clearly favored TIV in 97.8% of the trajectories and fr-TIV in only 0.5% of the cases. No trajectory favored TEIV as the best model. Consequently, for these short, sparsely sampled Ct-only trajectories, the simpler TIV model provided the most compelling, parsimonious description of within-host SARS-CoV-2 infection dynamics. This does not necessarily indicate the absence of eclipse-phase or memory-like dynamics in biological systems but highlights that the more complex models are not favored by standard information criteria for the limited information contained in these specific datasets.",
  "summary": "The main question in model selection for longitudinal RT-qPCR data is whether improvements in fit justify the additional parameters of more complex within-host models. We compared a target-cell-limited model (TIV), an eclipse-phase extension (TEIV), and a fractional-order TIV using infection waves reconstructed from SARS-CoV-2 cycle-threshold (Ct) measurements in a National Basketball Association…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}