Multi-omics Prediction from High-content Cellular Imaging with Deep Learning

High-content cellular imaging, transcriptomics, and proteomics data provide rich and complementary views on the molecular layers of biology that influence cellular states and function. However, the biological determinants through which changes in multi-omics measurements influence cellular morphology have not yet been systematically explored, and the degree to which cell imaging could potentially enable the prediction of multi-omics directly from cell imaging data is therefore currently unclear. Here, we address the question of whether it is possible to predict bulk multi-omics measurements directly from cell images using Image2Omics -- a deep learning approach that predicts multi-omics in a cell population directly from high-content images stained with multiplexed fluorescent dyes. We perform an experimental evaluation in gene-edited macrophages derived from human induced pluripotent stem cell (hiPSC) under multiple stimulation conditions and demonstrate that Image2Omics achieves significantly better performance in predicting transcriptomics and proteomics measurements directly from cell images than predictors based on the mean observed training set abundance. We observed significant predictability of abundances for 5903 (22.43%; 95% CI: 8.77%, 38.88%) and 5819 (22.11%; 95% CI: 10.40%, 38.08%) transcripts out of 26137 in M1 and M2-stimulated macrophages respectively and for 1933 (38.77%; 95% CI: 36.94%, 39.85%) and 2055 (41.22%; 95% CI: 39.31%, 42.42%) proteins out of 4986 in M1 and M2-stimulated macrophages respectively. Our results show that some transcript and protein abundances are predictable from cell imaging and that cell imaging may potentially, in some settings and depending on the mechanisms of interest and desired performance threshold, even be a scalable and resource-efficient substitute for multi-omics measurements. Overall design: For transcriptomic profiling we utilised a multiplexed 3' RNA-seq approach based on DRUG-seq (https://doi.org/10.1038/s41467-018-06500-x). 384-well plates of iPSC-monocyte precursors were subjected to genetic perturbation using CRISPR-Cas9 to knock out (KO) single genes per well or control conditions without CRISPR or non-targeting control gRNAs. Cells were then exposed to macrophage differentaiation stimuli before being stimulated with M0/M1/M2 skewing conditions with the exception of column 13 which was always M0 to provide a baseline control in each plate. Replicate groups in the supplementary csv files denominate triplicates of CRISPR gene targets per plate, positive control=drive stimulation phenotype, negative control= attenuate stimulation phenotype, neutral=no change in phenotype. Each 384-well plate results in 4x pairs of R1 and R2 reads due to amplification of the cDNA library with 4x pairs of primers.