Pledge Therapeutics reposted this
Our #PledgeAi team, represented by Nick Hamilton and Asanga Bandara, just presented our latest work at the AACR Journals Special Conference on AI in Cancer Research in #Montréal, bringing deep, experimentally rooted, structure-based drug design data we routinely generate into a field full of complexity, and, at times, more modeling than molecular reality. It’s hard not to reflect on how much AI is being asked to do in cancer research today: from interpreting spatial & -omics data to predicting treatment response, guiding drug combinations, stratifying patients, identifying new drug targets, and discovering new drugs. There’s promise here but also a lot of noise. And this classical AI challenge is compounded further in oncology research. The landmark 2019 study by Lin et al., published in Nature, provided a sobering reminder that over two-thirds (>66%) of approved #cancer drugs fail to act through their intended molecular targets, underscoring the fragility of our assumed “ground truths”. Assays often mislead, biological redundancy can obscure, and confound underlying principles. #Deeplearning models trained only on curated literature or retrospective screens tend to reinforce those blind spots unless they’re actively challenged with new, noisy, real-world data. #AI can be powerful, not as a shortcut to bypass experimentation, but to develop and test ideas faster, expose where our assumptions break, and focus experiments where they matter most. At Pledge Therapeutics, AI is one of many verticals we deploy. #PicoGen, our generative design platform, builds on years of experimental work to tackle a core problem in AI-driven drug design i.e., data scarcity. It unlocks a vast “undrugged” therapeutic space, by enabling de novo hit-generation against targets that currently have no known drug molecules. No where is this need more apparent than in cancer, where most disease drivers remain undrugged and unevaluated. PicoGen helps us generate hits for targets that currently have no Medicinal Chemistry starting points at all. More to come when we share our full AACR abstract. For now, here’s a glimpse of the work and the people behind it. #AACR #DrugDiscovery #CancerResearch #StructureBasedDesign #GenerativeAI #Biotech #AACRai25 #undruggable #oncology
-
-
-
-
-
+2