Generative De Novo Design
Anew’s Insight:
For challenging undruggable targets, the ligand space we need to explore is fundamentally larger and more complex than what we’re used to for canonical pockets.
Existing chemical HTS/VS libraries are biased toward canonical druggable pockets, making it a key challenge to find a proper hit for challenging targets.
Anew’s Solution:
Our mission is to develop a generative foundation model that learns atomic interactions, which is the universal language across diverse biomolecules.
Working with expertise in physics, chemistry, biology, and computer science, we aim to unlock the full potential of generative modeling, scaling across data and model sizes to understand the language of molecules.
AI-Physics Convergence
Unrivaled Affinity Prediction
A high-throughput platform delivering physics-based generalization superior to that of AI models and accuracy exceeding all open-source alternatives.
Target-Centric Parameterization
Data-driven parameterization rigorously optimized for complex drug molecules and challenging protein systems.
AI-Enhanced Sampling
Accelerating molecular dynamics to uncover rare conformational events beyond standard timescales.
Chemistry Agent
Precision Task Orchestration
Interprets objectives, matches toolchains, initiates tasks, manages data/dependencies/resource allocation.
Cross-platform Synergy
Integrates core biocomputing capabilities, enables multi-system collaboration and data interoperability.
Insight-Driven Analysis
Extracts metrics, runs statistical analysis, generates reports, supports one-click data export.
Knowledge-Embedded Support
Delivers workflow guidance, troubleshooting and document retrieval via built-in biocomputing knowledge base.