SOURCE-LINKED INTELLIGENCE

CellCourier: programmable drug delivery via combinatorial cell targeting and generative design of tailored binding peptides

CORDIS · observation · Publication date unknown

tides CellCourier aims to revolutionize mRNA delivery by enabling precise cell-type-specific targeting through a data-driven platform that integrates single-cell genomics, computational modeling, and generative AI. Conventional approaches often suffer from off-target effects, leading to unintended side effects. To address this, CellCourier leverages combinatorial peptide binder design to achieve precise cell-type recognition, enabling next-generation delivery modalities such as LNPs or STVs. At the core of CellCourier is a computational pipeline that defines unique surface codes for target cells using single-cell transcriptomic data from the Human Cell Atlas (HCA). This process is enhanced by Nicheformer, a model for predicting cell-cell interactions, and Prophet, a deep-learning model for multi-omics perturbation prediction—both key outcomes of the ERC AdG DeepCell grant. Using these refined surface codes, we apply our all-atom generative model to design peptide binders optimized for high specificity and stability. A competitive binding pipeline, which has shown a threefold increase

Read original source ↗ Open in workspace

recordType

award

status

SIGNED

region

EU

value

150000

unit

EUR