MolGene-E: Inverse Molecular Design to Modulate Single Cell Transcriptomics

Rahul Ohlan, Raswanth Murugan, Li Xie, Mohammadsadeq Mottaqi, Shuo Zhang, Lei Xie

Keywords: Systems Pharmacology, Drug Discovery, Phenotypic Screening, Generative AI, CLIP

Abstract: Designing drugs that can restore a diseased cell to its healthy state is an emerging approach in systems pharmacology to address medical needs that conventional target-based drug discovery paradigms have failed to meet. Single-cell transcriptomics can comprehensively map the difference between diseased and healthy cellular states, making it a valuable technique for systems pharmacology. However, single-cell omics data is highly noisy, heterogenous, scarce, and high-dimensional. As a result, no machine learning methods currently exist to use single-cell omics data to design new drug molecules. We have developed a new generative artificial intelligence (AI) framework named MolGene-E that can tackle this challenge. MolGene-E combines two novel models: 1) a cross-modal model that can harmonize and denoise chemical-perturbed data, and 2) A VAE-CLIP based generative model that can generate new drug molecules based on transcriptomics data. This makes it a potentially powerful new AI tool for drug discovery.

Submission Number: 188