Skewes-Cox, P. | Search bioRxiv

bioRxiv2024

A diffusion model conditioned on compound bioactivity profiles for predicting high-content images

High-content imaging (HCI) provides a rich snapshot of compound-induced phenotypic outcomes that augment our understanding of compound mechanisms in cellular systems. Generative imaging models for HCI provide a route towards anticipating the phenotypic outcomes of chemical perturbations in silico at unprecedented scale and speed. Here, we developed Profile-Diffusion (pDIFF), a generative method leveraging a latent diffusion model conditioned on in silico bioactivity profiles to predict high-content images displaying the cellular outcomes induced by compound treatment. We trained and evaluated a pDIFF model using high-content images from a Cell Painting assay profiling 3750 molecules with corresponding in silico bioactivity profiles. Using a realistic held-out set, we demonstrate that pDIFF provides improved predictions of phenotypic responses of compounds with low chemical similarity to compounds in the training set compared to generative models trained on chemical fingerprints only. In a virtual hit expansion scenario, pDIFF yielded significantly improved expansion outcomes, thus showcasing the potential of the methodology to speed up and improve the search for novel phenotypically active molecules.

bioinformatics↗

bioRxiv2024

Tissue damage during acute Trypanosoma cruzi infection is associated with reduced reparative regulatory T cell response and can be attenuated by early interleukin-33 administration.

Tissue-repair regulatory T cells (trTregs) constitute a specialized regulatory subset renowned for orchestrating tissue homeostasis and repair. While extensively investigated in sterile injury models, their role in infection-induced tissue damage and the regulation of protective antimicrobial immunity remains largely unexplored. This investigation examines trTregs dynamics during acute Trypanosoma cruzi infection, a unique scenario combining extensive tissue damage with robust antiparasitic CD8+ immunity. Contrary to conventional models of sterile injury, our findings reveal a pronounced reduction of trTregs in secondary lymphoid organs and tissues during acute T. cruzi infection. This unexpected decline correlates with systemic as well local tissue damage, as evidenced by histological alterations and downregulation of repair-associated genes in skeletal muscle. Remarkably, a parallel decrease in systemic levels of IL-33, a crucial factor for trTregs survival and expansion, was detected. We found that early treatment with systemic recombinant IL-33 during infection induces a notable surge in trTregs, accompanied by an expansion of type 2 innate lymphoid cells and parasite-specific CD8+ cells. This intervention results in a mitigated tissue damage profile and reduced parasite burden in infected mice. These findings shed light on trTregs biology during infection-induced injury and demonstrate the feasibility of enhancing a specialized Tregs response without impairing the magnitude of effector immune mechanisms, ultimately benefiting the host. Furthermore, this study settles groundwork of relevance for potential therapeutic strategies in Chagas disease and other infections. AUTHOR SUMMARYChagas disease, caused by the protozoan Trypanosoma cruzi, induces severe organ damage caused by the interplay between the parasite and the immune response. In our investigation, we delved into the role of tissue-repair regulatory T cells (trTregs) during the acute phase of T. cruzi infection in mice. Surprisingly, we observed a reduction in trTregs during the peak of tissue damage, contrary to their usual accumulation after injury in other contexts. This decline aligned with decreased levels of interleukin-33, a critical factor for trTregs survival. Administering interleukin-33 at early infection times not only boosted trTregs but also expanded other reparative and antiparasitic immune cells. Consequently, these treated mice exhibited reduced damage and lower parasite levels in tissues. Our findings offer insights into trTregs behavior during infection-induced injury, suggesting a promising avenue for therapeutic interventions in Chagas disease and related conditions. This study lays the groundwork for potential strategies that balance the immune response, supporting tissue repair without compromising the ability to control the infection, which could have broader implications for infectious diseases and tissue damage-related pathologies.

immunology↗

bioRxiv2023

Compound activity prediction with dose-dependent transcriptomic profiles and deep learning

Predicting compound activity in assays is a long-standing challenge in drug discovery. Computational models based on compound-induced gene-expression signatures from a single profiling assay have shown promise towards predicting compound activity in other, seemingly unrelated, assays. Applications of such models include predicting mechanisms-of-action (MoA) for phenotypic hits, identifying off-target activities, and identifying polypharmacologies. Here, we introduce Transcriptomics-to-Activity Transformer (TAT) models that leverage gene-expression profiles observed over compound treatment at multiple concentrations to predict compound activity in other biochemical or cellular assays. We built TAT models based on gene-expression data from a RASL-Seq assay to predict the activity of 2,692 compounds in 262 dose response assays. We obtained useful models for 51% of the assays as determined through a realistic held-out set. Prospectively, we experimentally validated the activity predictions of a TAT model in a malaria inhibition assay. With a 63% hit rate, TAT successfully identified several sub-micromolar malaria inhibitors. Our results thus demonstrate the potential of transcriptomic responses over compound concentration and the TAT modeling framework as a cost-efficient way to identify the bioactivities of promising compounds across many assays.

bioinformatics↗