Quantum Machine Learning Algorithms for Drug Discovery Applications.

Area of Science:

• Computational chemistry and cheminformatics
• Quantum machine learning applications
• Drug discovery and development

Background:

• Vast datasets of small molecules screened against biological targets are available for drug discovery.
• Classical machine learning algorithms like Support Vector Machines (SVM) and Deep Neural Networks (DNN) are computationally intensive for large datasets.
• Quantum computing (QC) algorithms show promise for accelerating machine learning but face limitations, particularly in data compression for cheminformatics.

Purpose of the Study:

• To demonstrate a method for compressing large molecular descriptor datasets for quantum computing applications in drug discovery.
• To benchmark quantum Support Vector Machines (SVM) and data reuploading classifiers against classical and hybrid approaches.
• To illustrate the necessary steps for achieving 'quantum computer readiness' in cheminformatics for drug discovery.

Main Methods:

• Application of Support Vector Machines (SVM) and a data reuploading classifier (a Deep Neural Network equivalent) on a quantum computer.
• Compression of molecular descriptor data for datasets ranging from hundreds to hundreds of thousands of molecules.
• Benchmarking quantum approaches against classical computer (CC) and hybrid computational methods.

Main Results:

• Successful compression of large molecular datasets for quantum machine learning algorithms.
• Demonstration of quantum algorithms' performance on datasets relevant to SARS-CoV-2, plague, and *M. tuberculosis* screening.
• Validation of quantum approaches in comparison to classical and hybrid methods.

Conclusions:

• The study provides a foundational framework for applying quantum computing to drug discovery through effective data compression.
• Achieving 'quantum computer readiness' is crucial for leveraging QC's potential in accelerating drug discovery pipelines.
• This work paves the way for future advancements in quantum-enhanced cheminformatics and drug discovery.