1910 offers ITO Platform
ProductOperating Status: Currently Offered
1910's ITO™ platform orchestrates multimodal AI and lab automation for small and large molecule drug discovery.
1910's ITO™ platform orchestrates multimodal AI and lab automation for small and large molecule drug discovery.
ITO joins target identification, molecule design, and laboratory validation in one stated workflow. The provider describes data orchestration, a collection of AI models, molecular synthesis planning, and robotics-driven assays rather than a model-only discovery tool.
Its federated-learning architecture allows pharmaceutical partners to train a shared model while retaining data locally. The provider also names its Microsoft Azure Quantum Elements partnership and GPU infrastructure as support for its computing workflow.
The platform's claimed discovery workflow still requires manufactured candidates and wet-lab assays before a designed molecule can be evaluated. Its own description places biochemical, cellular, pharmacological, and potentially animal-model work after in-silico design.
The federated-learning option depends on participating pharmaceutical partners training local models and returning model updates.
1910 describes ITO as its current input-transform-output platform for target identification and molecular design. Its first-party platform page connects data orchestration, a model foundry, automated laboratory validation, and federated learning under that name.
The inspected platform page does not state when ITO was first introduced or released. A year for the company or a date for one later feature would not establish the platform's own origin, so its first-release year remains unverified.
1910's ITO platform combines multimodal biological and computational data, AI models, and automated laboratory work for drug discovery. Its input stage prepares patient, assay, sequencing, and simulation data; its model system then supports target identification and molecular design.
The platform addresses both small and large molecules. It proposes synthetic routes for computationally designed candidates and uses biochemical, cellular, and pharmacological assays to test manufactured molecules, while a federated-learning option lets partners train models without sharing their underlying datasets.