bioq: a unified, agent-native command-line interface to a fleet of AI drug-discovery methods
bioq is an open-source, lightweight command-line interface that unifies access to a fleet of 38+ containerized AI drug-discovery tools via a serverless gateway, enabling researchers and automated agents to execute complex workflows from a laptop without managing local hardware or conflicting software dependencies.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of a preprint that has not been peer-reviewed. It is not medical advice. Do not make health decisions based on this content. Read full disclaimer
The search for new medicines has long been a slow, expensive journey, often taking years and billions of dollars to bring a single treatment from a laboratory idea to a pharmacy shelf. In recent years, artificial intelligence has begun to accelerate this process, offering powerful tools that can predict how a molecule will fold into a shape, design entirely new chemical structures from scratch, or estimate how well a drug might bind to a disease target. These methods cover every stage of the discovery pipeline, from the initial design of a molecule to checking if it will be safe for the human body. However, a significant barrier has emerged: while these tools are individually powerful, they are built to run on specific, often conflicting computer systems. Most require complex software setups and access to massive, expensive computer hardware that only large pharmaceutical companies or well-funded research centers can afford. This fragmentation means that a scientist cannot easily link these tools together to run a full experiment, because the software required for one step often breaks the software needed for the next.
To solve this problem, a team of researchers has introduced a new system called bioq, which acts as a unified command line for a vast collection of these artificial intelligence drug-discovery tools. Instead of forcing researchers to install dozens of different, incompatible programs on their own computers, bioq provides a single, consistent interface that connects to a fleet of specialized services running in the cloud. The system is designed to be lightweight and easy to use, requiring no local installation of complex graphics processing software or deep knowledge of cloud computing configurations. A researcher or an automated computer agent can simply type a command into a standard terminal on a personal laptop to access any of the available tools. The heavy lifting happens remotely on server-based computers that charge only for the time the job takes to run, removing the need for expensive, dedicated hardware in the lab.
The core of this system is a gateway that manages a growing fleet of more than thirty-eight different containerized tools, each representing a specific step in the drug discovery process. These tools cover seven distinct stages of discovery and can handle six different types of molecular data. Because the interface is self-describing and consistent, the same basic commands work regardless of which specific tool is being used. This uniformity allows researchers to chain multiple steps together into a single workflow without worrying about whether the software environments will clash. The system is built on open-source software, meaning the code is freely available for anyone to inspect, modify, or host on their own servers if they prefer. The researchers have provided clear instructions for setting up the system locally or using it through public cloud services, ensuring that the technology is accessible to anyone with a standard internet connection and a computer running a modern version of Python.
By removing the technical friction of software installation and hardware requirements, bioq transforms the landscape of computational drug discovery. It shifts the focus from managing complex computer environments to simply running experiments. The system is designed to serve as a foundation for automated, agent-driven discovery, where computer programs can independently plan and execute complex research tasks without human intervention at every step. The project is actively maintained, with the researchers committed to adding new tools and features over time. This approach does not claim to have solved the biological mysteries of disease, but it offers a practical, accessible way for scientists and automated agents to utilize the full power of modern artificial intelligence tools, regardless of the size of their laboratory or their budget for computer hardware.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.