Utilizing contrastive machine learning models to extract joint-space molecular-phenomic embeddings from molecular structures or phenomic images

Abstract

The present disclosure relates to systems, non-transitory computer-readable media, and methods for utilizing a contrastive molecular-phenomic embedding model that learns joint latent space embeddings between molecular structures and phenomic images to generate molecular-phenomic embeddings that represent molecular impacts on cellular functions. Indeed, the disclosed systems can utilize phenomic image embeddings generated from a pretrained phenomic image encoder model and corresponding molecular structural embeddings with a contrastive molecular-phenomic embedding model to learn a joint latent space between molecular structures and phenomic images (with an inter-sample similarity aware loss (S2L), explicit and implicit concentration encoding, and under sampling of inactive molecular data). In addition, the disclosed systems can utilize molecular structures and/or phenomic images with the contrastive molecular-phenomic embedding model to generate molecular-phenomic embeddings that enable a variety of molecular inferences (e.g., similar molecule determinations, similar phenomic image determinations, phenotypic impact determinations from particular molecules, and/or molecular activity classifications).