THANK YOU FOR SUBSCRIBING
Pharma Tech Outlook | Wednesday, April 17, 2024
Artificial intelligence and machine learning allow scientists to spend more time designing new experiments or scaling up existing workflows.
FREMONT, CA: Automation goals once revolved around getting more done. The goal is to build a smart laboratory where multiple instruments are connected and controlled by a central platform. By capturing and storing all the data generated, an information management system makes it easier for ML algorithms to detect patterns and make smarter decisions.
Stay ahead of the industry with exclusive feature stories on the top companies, expert insights and the latest news delivered straight to your inbox. Subscribe today.
Here are a few trends in laboratory automation, from AI-driven drug discovery to robot scientists.
Workflows with end-to-end automation: End-to-end automation has evolved from simply performing repetitive tasks to intuitively fulfilling the entire workflow from start to finish. An automated facility will execute workflows autonomously if different systems are connected and communicate. Cell-based microscopy uses a centrally connected liquid handler, sample mover, incubator, and microscope. When the sequence is triggered, the workflow initiates, advancing from one step to another while capturing data and sending it to a cloud-based server. These systems are linked via special software, which plans workflow sequences, monitors progress remotely, sometimes in real time, and connects them all. By integrating an ecosystem of automation, white space between individual workflows can be significantly eliminated, leading to a more efficient, smoother laboratory operation.
Centralizing data and adopting good data practices: The vast amount of data generated by automation is subject to the data practices enforced within the laboratory. A digital infrastructure is needed to manage the large amount of data generated by automated workflows. It is necessary to capture, annotate, and store the data generated by these workflows appropriately. Instruments, samples, inventory, and end-to-end workflows can typically be centralized and managed using digital tools. An application programming interface (API) connects instruments with their vendor software platforms to the central data repository.
Leveraging AI in small and big ways: According to a report from 2021, AI will dominate the pharmaceutical industry in the coming years, with nearly 100 collaborations between pharmaceutical companies and AI vendors in recent years. AI is already successfully used in the tech industry and for image and speech recognition, but it is now also used in drug discovery. The data from a series of laboratory experiments with automated workflows can be used to train AI systems in the 'rules' of the experiment and generate new knowledge based on them. In essence, AI refers to the ability of a computer to simulate human cognitive functions, such as learning and troubleshooting. Through data, the computer learns and improves without being intervened, imparting intelligence to it.
More in News