Drug discovery traditionally involves collaboration among scientists and engineers from various disciplines. AI's recent role in drug discovery demands increased integration of AI/ML experts, data scientists, and AI/ML Ops. The platform offers AI/ML models crucial for efficient drug discovery.

Previously, scientists faced challenges in analyzing target protein structures to identify drug-design pockets, often leading to failed discovery processes. Our platform addresses this with an AI-driven active site identification module, enabling users to find active sites without prior protein knowledge for in silico drug discovery.

In silico docking screening provides binding energy in Kcal/mol, requiring biologists to determine ligand concentrations for experimental tests. To simplify this, the platform incorporates an AI model estimating pharmacokinetics values in molar concentration, streamlining experimental setup.

Additionally, our platform features an AI model for ADME-Tox predictions. Unlike others, it integrates with Protein-Ligand docking results, enabling quick identification and modification of toxic functional groups without compromising binding energies or efficacies.

Previously, CPU-based docking processes limited efficiency, despite AI inference being GPU-compatible. The platform introduces dynamic GPU allocation, allowing users to add more resources and efficiently share multi-GPU setups in real-time.

Our SaaS platform enables users to discover superior molecules in less time and cost. Its pay-as-you-go model avoids yearly subscriptions and large server expenses, empowering users to maintain a competitive edge by paying only for utilized resources.

Previously drug discovery has been an orchestration of multi-disciplinary works involving many scientists and engineers ranging from biochemists, structural biologists, medicinal chemists, and bioinformaticians. The latest development of AI in drug discovery requires more demanding workforce integration like AI/ML experts for hyperparameter optimization Data scientists for feature engineering and AI/ML Ops for training server management.

This company can build and launch a very cost-effective platform with a multi-disciplinary team for drug discovery.

As a result, SaaS platform was developed to assist scientists or companies who want to integrate AI into their drug discovery pipeline. Unlike licensing based installable software, this platform utilizes Web3.0 standard and fully functional on a web-browser and it charges pay-as-you-go. So, users don’t have to install and maintain their server anymore. Inside the platform users can drag-and-connect each AI/ML model as if it is a node and edges so the computation can be executed as a graph node. In this way a user can increase his/her creative potentials to find candidate molecules smarter.

- Type of partner sought

Companies, public research institutes, and university research institutes that are related to the development of new drugs or functional material.

- Specific area of activity of the partner

The pharmaceutical industry, the functional food industry, Natural compound research area (industry, university, etc)