Search papers, labs, and topics across Lattice.
DECODE, a novel framework, bridges the gap between structural and phenotypic screening by empowering chemical representations with biological semantics. It leverages paired transcriptomic and morphological data as supervisory signals to extract measurement-invariant biological fingerprints from chemical structures while filtering experimental noise. DECODE demonstrates a 20% relative improvement in mechanism-of-action (MOA) prediction and a 6-fold increase in hit rates for novel anti-cancer agents in zero-shot settings.
Unlock a 6x boost in hit rates for novel anti-cancer agents by infusing chemical structures with biological insights, even without biological data at inference time.
Motivation: The scalable identification of bioactive compounds is essential for contemporary drug discovery. This process faces a key trade-off: structural screening offers scalability but lacks biological context, whereas high-content phenotypic profiling provides deep biological insights but is resource-intensive. The primary challenge is to extract robust biological signals from noisy data and encode them into representations that do not require biological data at inference. Results: This study presents DECODE (DEcomposing Cellular Observations of Drug Effects), a framework that bridges this gap by empowering chemical representations with intrinsic biological semantics to enable structure-based in silico biological profiling. DECODE leverages limited paired transcriptomic and morphological data as supervisory signals during training, enabling the extraction of a measurement-invariant biological fingerprint from chemical structures and explicit filtering of experimental noise. Our evaluations demonstrate that DECODE retrieves functionally similar drugs in zero-shot settings with over 20% relative improvement over chemical baselines in mechanism-of-action (MOA) prediction. Furthermore, the framework achieves a 6-fold increase in hit rates for novel anti-cancer agents during external validation. Availability and implementation: The codes and datasets of DECODE are available at https://github.com/lian-xiao/DECODE.
