Cradle

Cradle provides an AI-powered protein design platform specifically built for scientists and research teams in biotechnology and pharmaceutical industries. The platform enables users to engineer proteins with improved properties—such as stability, activity, binding affinity, or expression yield—by leveraging machine learning models trained on both public and proprietary experimental data. Researchers can generate novel protein candidates directly from sequence or structure inputs, then track each candidate through an iterative design-build-test-learn cycle. The system supports uploading experimental results, visualizing round status, and managing large sets of sequences and associated metadata in a structured dashboard. A key function is the ability to incorporate the user’s own experimental data as feedback, allowing the AI models to refine predictions over successive rounds. This closed-loop approach reduces the number of wet-lab experiments needed to achieve desired protein characteristics, saving time and resources. The platform also includes features for exploring reports, organizing projects by round or property target, and collaborating within teams. Cradle is used by top biopharma and industrial bio R&D organizations—such as Corteva and Lundbeck—to accelerate discovery pipelines for enzyme engineering, therapeutic protein optimization, and industrial biocatalyst development. The algorithms can handle diverse protein classes, including antibodies, enzymes, and de novo designed polypeptides. Specific tasks supported include predicting mutational effects, designing libraries with targeted diversity, and identifying sequences that meet multiple constraints simultaneously (e.g., high thermostability plus catalytic activity). The platform is accessed via a web interfac