Abstract
Existing drug discovery pipelines take 5-10 years and cost billions of dollars. Computational approaches aim to sample from regions of the whole molecular and solid-state compounds called chemical space which could be on the order of 1060 . Deep generative models can model the underlying probability distribution of both the physical structures and property of drugs and relate them nonlinearly. By exploiting patterns in massive datasets, these models can distill salient features that characterize the molecules. Generative Adversarial Networks (GANs) discover drug candidates by generating molecular structures that obey chemical and physical properties and show affinity towards binding with the receptor for a target disease. However, classical GANs cannot explore certain regions of the chemical space and suffer from curse-of-dimensionality. A full quantum GAN may require more than 90 qubits even to generate QM9-like small molecules. We propose a qubit-efficient quantum GAN with a hybrid generator (QGAN-HG) to learn richer representation of molecules via searching exponentially large chemical space with few qubits more efficiently than classical GAN. The QGANHG model is composed of a hybrid quantum generator that supports various number of qubits and quantum circuit layers, and, a classical discriminator. QGAN-HG with only 14.93% retained parameters can learn molecular distribution as efficiently as classical counterpart. The QGAN-HG variation with patched circuits considerably accelerates our standard QGANHG training process and avoids potential gradient vanishing issue of deep neural networks. Code is available on GitHub https://github.com/jundeli/quantum-gan.
Highlights
The drug development pipeline consists of stages of target discovery, molecular design, preclinical studies, and clinical trials, which makes the process of creating a marketable drug expensive and time consuming [1]
Searching new drugs can be considered as navigating through the chemical space, which is an ensemble of all organic molecules
Quantum GAN with hybrid generator (QGAN-HG) RESULTS Fig. 3 shows QGAN-HG performance may rely on both qubit count N and repeatable layer count L
Summary
The drug development pipeline consists of stages of target discovery, molecular design, preclinical studies, and clinical trials, which makes the process of creating a marketable drug expensive and time consuming [1]. We, (1) propose a novel quantum GAN mechanism with hybrid generator to qubit-efficiently tackle any real-world learning tasks solvable on classical GANs; (2) generate drug molecular graphs with quality similar to classical methods in terms of Frchet Distance (FD) score and drug property scores, and achieve high training efficiency with less than 20% of the original parameters; (3) design a quantum GAN with patched sub-circuits which significantly decreases training time, making the quantum algorithm efficiently executable even in simulation environment; (4) provide a new quantum paradigm for generative models which bypass the thorny training instability issues; (5) validate the capability of generating small drug molecules on real IBM quantum computers by running inference stage of QGAN-HG model
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
