Additive Dose Response Models: Explicit Formulation and the Loewe Additivity Consistency Condition.

Area of Science:

• Pharmacology
• Computational Biology
• Drug Discovery

Background:

• High-throughput screening enables massive drug combination analysis.
• Identifying drug interactions requires comparing combinations to a baseline of no interaction.
• Loewe Additivity is a common principle for defining this baseline, assuming no self-interaction and dose equivalence.

Purpose of the Study:

• To derive and evaluate explicit and implicit null reference models for drug interactions based on Loewe Additivity.
• To investigate the Loewe Additivity Consistency Condition (LACC) and its impact on model predictions.
• To compare the performance of the novel explicit model against the established implicit model using real-world drug screening data.

Main Methods:

• Derivation of explicit and implicit null reference models from Loewe Additivity.
• Introduction and analysis of the Loewe Additivity Consistency Condition (LACC).
• Evaluation of models using two non-interactive drug screening datasets (Cokol et al., 2011; Yadav et al., 2015).
• Comparison of model bias and mean squared error.

Main Results:

• The Loewe Additivity Consistency Condition (LACC) is frequently violated in practice.
• Violated LACC leads to divergent predictions between explicit and implicit models.
• The novel explicit null reference model demonstrated smaller mean squared errors compared to the implicit model.
• The explicit model offers significantly faster computation.

Conclusions:

• The LACC is a critical factor in accurately defining Loewe Additivity.
• The explicit null reference model provides a more robust and efficient approach for analyzing drug interactions.
• This work offers improved methods for high-throughput drug screening analysis.