Track the frontier, from condensed matter to cosmology.
Search papers, labs, and topics across Lattice.
This paper introduces a hybrid classical-quantum neural network architecture for predicting receptor functional potency, specifically focusing on GLP-1R cAMP-related pEC50. By jointly training a classical molecular encoder with a linear base head and a shallow quantum-circuit residual head, the models achieve improved predictive accuracy, reducing mean absolute error by approximately 5.6% on a structure-group split for GLP-1R. The study further suggests that the observed gains may stem from the quantum residual's ability to capture structured quadratic feature interactions, informing future compact residual-head designs.
Quantum-circuit residuals boost receptor functional potency prediction accuracy by capturing structured quadratic feature interactions, outperforming classical baselines.
The predictive performance of neural networks for receptor functional potency modeling depends on both molecular representations and prediction-head design. Using GLP-1R cAMP-related pEC50 prediction as the primary task, we jointly train a classical molecular encoder with a linear base head and a shallow quantum-circuit residual head that share the learned molecular representation. We evaluate the models using random and structure-group splits, include tuned Chemprop as a domain reference, and extend the evaluation to ADRB2 and MC4R. Under the GLP-1R structure-group split, the quantum-residual models reduce mean absolute error relative to the corresponding base models from 0.677 to 0.639 with a multilayer perceptron (MLP) encoder and from 0.680 to 0.641 with a graph neural network (GNN) encoder. In structural control experiments, a parameter-count-matched real-orthogonal control achieves similar predictive performance, suggesting that some quantum-residual gains may be associated with structured quadratic feature interactions. By combining receptor functional potency prediction with structural analysis of the quantum residual, this study provides empirical evidence to inform compact residual-head design and highlights the potential of quantum-circuit parameterizations for this task.