Understanding cellular responses to stimuli is crucial for biological discovery and drug development. Transcriptomics provides interpretable, gene-level insights, while microscopy imaging offers rich predictive features but is harder to interpret. Weakly paired datasets, where samples share biological states, enable multimodal learning but are scarce, limiting their utility for training and multimodal inference. We propose a framework to enhance transcriptomics by distilling knowledge from microscopy images. Using weakly paired data, our method aligns and binds modalities, enriching gene expression representations with morphological information. To address data scarcity, we introduce (1) Semi-Clipped, an adaptation of CLIP for cross-modal distillation using pretrained foundation models, achieving state-of-the-art results, and (2) PEA (Perturbation Embedding Augmentation), a novel augmentation technique that enhances transcriptomics data while preserving inherent biological information. These strategies improve the predictive power and retain the interpretability of transcriptomics, enabling rich unimodal representations for complex biological tasks.

Deriving insights from experimentally generated datasets requires methods that can account for random and systematic measurement errors and remove them in order to accurately represent the underlying effects of the conditions being tested. Here we present a framework for pretraining on large-scale microscopy datasets that includes three steps: (1) curating a set of diverse and self-consistent training samples, (2) scaling training of an appropriate foundation model architecture on this dataset, (3) evaluating intermediate layers of the trained model to identify the best representation for downstream tasks. Using this strategy, we present the largest foundation model for cell microscopy data to our knowledge, a new 1.9 billion-parameter ViT-G/8 MAE trained on over 8 billion microscopy image crops. Compared to a previous published ViT-L/8 MAE, our new model achieves a 60% improvement in linear separability of genetic perturbations and obtains the best overall performance on whole-genome relationship recall, batch correction replicate consistency, and compound-gene activity prediction benchmarks.

Unlike color photography images, which are consistently encoded into RGB channels, biological images encompass various modalities, where the type of microscopy and the meaning of each channel varies with each experiment. Im-portantly, the number of channels can range from one to a dozen and their correlation is often comparatively much lower than RGB, as each of them brings specific information content. This aspect is largely overlooked by methods designed out of the bioimage field, and current solutions mostly focus on intra-channel spatial attention, often ignoring the relationship between channels, yet crucial in most biological applications. Importantly, the variable channel type and count prevent the projection of several experiments to a unified representation for large scale pre-training. In this study, we propose ChAda-ViT, a novel Channel Adaptive Vision Transformer architecture employing an Inter-Channel Attention mechanism on images with an arbitrary number, order and type of channels. We also introduce IDR-CellI 00k, a bioimage dataset with a rich set of 79 experi-ments covering 7 microscope modalities, with a multitude of channel types, and counts varying from 1 to 10 per exper-iment. Our architecture, trained in a self-supervised man-ner, outperforms existing approaches in several biologically relevant downstream tasks. Additionally, it can be used to bridge the gap for the first time between assays with differ-ent microscopes, channel numbers or types by embedding various image and experimental modalities into a unified biological image representation. The latter should facilitate interdisciplinary studies and pave the way for better adoption of deep learning in biological image-based analyses.

The ability to predict the performance of a query before results are returned has been a longstanding challenge in Information Retrieval (IR) systems. Inspired by this task, we introduce, in this paper, a novel task called “Prompt Performance Prediction” that aims to predict the performance of a prompt, before obtaining the actual generated images. We demonstrate the plausibility of our task by measuring the correlation coefficient between predicted and actual performance scores across: three datasets containing pairs of prompts and generated images and three art domain datasets of real images and real user appreciation ratings. Our results show promising performance prediction capabilities, suggesting potential applications for optimizing user prompts.