Quantum Computing-Enhanced Algorithm Unveils Novel Inhibitors for KRAS
(2402.08210)Abstract
The discovery of small molecules with therapeutic potential is a long-standing challenge in chemistry and biology. Researchers have increasingly leveraged novel computational techniques to streamline the drug development process to increase hit rates and reduce the costs associated with bringing a drug to market. To this end, we introduce a quantum-classical generative model that seamlessly integrates the computational power of quantum algorithms trained on a 16-qubit IBM quantum computer with the established reliability of classical methods for designing small molecules. Our hybrid generative model was applied to designing new KRAS inhibitors, a crucial target in cancer therapy. We synthesized 15 promising molecules during our investigation and subjected them to experimental testing to assess their ability to engage with the target. Notably, among these candidates, two molecules, ISM061-018-2 and ISM061-22, each featuring unique scaffolds, stood out by demonstrating effective engagement with KRAS. ISM061-018-2 was identified as a broad-spectrum KRAS inhibitor, exhibiting a binding affinity to KRAS-G12D at $1.4 \mu M$. Concurrently, ISM061-22 exhibited specific mutant selectivity, displaying heightened activity against KRAS G12R and Q61H mutants. To our knowledge, this work shows for the first time the use of a quantum-generative model to yield experimentally confirmed biological hits, showcasing the practical potential of quantum-assisted drug discovery to produce viable therapeutics. Moreover, our findings reveal that the efficacy of distribution learning correlates with the number of qubits utilized, underlining the scalability potential of quantum computing resources. Overall, we anticipate our results to be a stepping stone towards developing more advanced quantum generative models in drug discovery.
Overview
-
This study introduces a hybrid quantum-classical generative model for the acceleration of therapeutic molecule identification, specifically targeting the KRAS G12D mutation in cancer therapy.
-
The model utilizes a Quantum Circuit Born Machine (QCBM) and Long Short-Term Memory (LSTM) networks to efficiently navigate molecular space and generate high-quality candidates for further testing.
-
Experimental validation through Surface Plasmon Resonance (SPR) and cell-based assays identified two promising compounds, demonstrating the model's potential in producing viable therapeutic candidates.
-
The research highlights the promise of quantum-computing enhancements in drug discovery and emphasizes the need for further investigations into the scalability and practical applications of such methodologies.
Advances in Quantum Assisted Drug Discovery: Targeting KRAS G12D
Introduction
In the realm of drug discovery, the integration of quantum computing with classical machine learning offers a promising avenue for accelerating the identification of potential therapeutic molecules. This study presents a hybrid quantum-classical generative model designed to enhance the discovery of novel ligands for the KRAS G12D mutation, a challenging target in cancer therapy. Employing a Quantum Circuit Born Machine (QCBM) coupled with Long Short-Term Memory (LSTM) networks, the research aims to leverage the strengths of both quantum and classical computing for drug design.
Quantum-Classical Hybrid Approach
The cornerstone of this research lies in the development and implementation of a quantum-classical hybrid algorithm. The model combines the generative capabilities of a QCBM with the sequential data processing strength of LSTM networks. This amalgamation aims to navigate the vast chemical space efficiently, generating high-quality molecular candidates for the KRAS G12D protein target.
Quantum Generative Model
At the quantum forefront, the QCBM serves as a sophisticated tool for sampling the molecular space. It utilizes parameterized quantum states and adheres to the Born rule for probability distributions. The model's training involves minimizing the Exact Negative Log-Likelihood (Exact NLL), an iterative process of adjusting quantum circuit parameters to approximate the target distribution accurately.
Classical Model Integration
Integrating LSTM into the workflow complements the quantum model by processing the sampled states to generate ligand structures. The LSTM network, known for its efficacy in handling sequential and temporal data, adapts the quantum-generated samples into realistic molecular designs. The validation of these designs employs a computational filter, Chemistry42, assessing ligand quality specifically for KRAS G12D interactions.
Experimental Evaluation
The empirical phase assessed the biological efficacy of generated ligands through Surface Plasmon Resonance (SPR) and cell-based assays. From numerous synthesized candidates, two compounds, ISM061-018-2 and ISM061-22, demonstrated significant promise. Both compounds showcased novel chemotypes and pronounced activity against KRAS G12D, underscoring the potential of the proposed hybrid model in generating viable therapeutic candidates.
Conclusions and Future Directions
This study underscores the substantial promise of quantum-enhanced methodologies in drug discovery. Despite the initial findings, it is crucial to recognize the challenge of establishing a quantum advantage over classical methods. The scalability and practical applications of quantum resources in drug discovery remain a pivotal area for future investigation. As quantum computing technology advances, its integration with classical algorithms is expected to play a transformative role in the pursuit of new therapeutics.
By bridging the gap between quantum computing and traditional drug discovery processes, this research paves the way for developing more sophisticated and efficient algorithms. Such advancements will undoubtedly contribute to the expedited discovery of novel drug candidates, with the potential to address some of the most elusive targets in medical science.
Create an account to read this summary for free: