Quantifying the "Gut Feeling": A New Framework for Tolerable Uncertainty in Read-Across

For years, expert toxicologists have relied on a mix of data and "professional intuition" to justify read-across. But in a regulatory landscape demanding more transparency, "trust me, I'm an expert" is no longer enough.

We are excited to highlight a seminal new paper: "A scheme for the assessment and definition of tolerable uncertainty in read-across for toxicological data gap filling" (Regulatory Toxicology and Pharmacology, 2026).

The authors, Mark Cronin and Terry Schultz, are names that need no introduction to this community. This paper is the culmination of decades of work.

Why You Should Read It

The paper moves us from subjective similarity to an ordinal scale of uncertainty. Key highlights include:

• The Seven Factors: A breakdown of uncertainty across metrics of chemical similarity, toxicodynamics (MIEs), toxicokinetics (metabolism/bioavailability), and data quality.

• The "Tolerable" Benchmark: Using real-world accepted REACH dossiers to define what level of uncertainty (Very Low to Moderate) regulators actually find acceptable.

• Endpoint Specificity: Why "similarity" for skin sensitization (protein reactivity) looks different than "similarity" for 90-day systemic toxicity (metabolic clearance).

Integrating Science into Training

At ToxNavigation, we believe that understanding why a model works is as important as the result itself. This paper's methodology is perfectly aligned with our flagship course:

"NAMs - Use and Application of QSAR and Read-Across"

We are currently updating the course modules to include this new uncertainty matrix. We will detail exactly how to generate the parameters required to fill the seven uncertainty areas identified by Cronin and Schultz.