Highlights
DeepCure and CWX demonstrate the potential for chemistry LLMs to increase productivity in medicinal chemistry. Real-time patent searches, publications, images, and language translation, combined with a custom Gemini AI chatbot created by CWX, enable scientists to interact with AI and receive reasoned, data-backed responses.
The challenge
Traditional methods for small molecule discovery were inefficient, requiring manual data extraction and analysis. DeepCure needed an AI-powered platform to speed up research and identify hidden chemical patterns to discover new drug candidates faster.
The solution
DeepCure worked with CWX to build AI models using Vertex AI and Document AI. These models automated data processing and harmonization, reducing manual efforts by 7x, and empowered DeepCure's scientists to accelerate their research and discovery process.
The implementation process
AI Model Development: Built custom models with Vertex AI to analyze chemical data at scale.
Platform Setup: Used GCP infrastructure for secure, scalable data storage.
Process Optimization: Streamlined data harmonization, cutting manual tasks by 7x.
Accelerated Research: Enabled faster iteration, speeding up drug candidate identification.
