Applying AI and Cloud Technologies for Multivariate Control in Upstream Processes

Applying AI and Cloud Technologies for Multivariate Control in Upstream Processes

Two unrecognizable pharmacologists wearing protective lab coats, masks and gloves working together on new vaccine

SARS-CoV-2, also known as COVID-19, has transformed the world. It has changed our perception of the frailty of human beings when faced with a well-designed, small combination of proteins and genetic material. We realize that a simple virus really can paralyze the world, affecting all aspects of life including social, economic, traveling, and general freedom to move about. Given these circumstances, many pharmaceutical companies are deriving their existing drugs hoping to minimize complications to implement strategies to curtail its expansion and to generate a future vaccine.

Artificial Intelligence (AI) has played a role in the COVID-19 crisis in many different ways. AI has been used to search for different treatments and vaccine approaches, to analyze the data from organ scans, to observe and anticipate the evolution of the pandemic worldwide based on specific indicators and AI has been applied as a decision-maker tool in public sectors. However, AI can also play a major role in the manufacturing of the new vaccines or Advanced Therapies (ATMP) where limited knowledge is available. The variability of biological entities is highly present and there is a public need and pressure to deliver a vaccine as soon as possible.

Most vaccine process development using the SARS-CoV-2 virus (or parts of its capsid) requires a specific process development which usually involves the expansion of a specific cell line and the determination of the ideal conditions for the generation of the vaccine. In this sense, AI and advanced analytics can help at different stages: batch information can be used to perform a risk analysis, discover the key critical factors impacting the process and enhance the continuous generation of process knowledge. In addition, the use of predictive models can help manufacturers to early detect possible deviations and react as soon as possible to fix anomalies during the process. Importantly, AI and real-time capabilities require GxP compliance tools that integrate the information in a contextualized manner. Cloud technologies provide a tremendous advantage for retrieving information in real-time from different equipment and manufacturing lines and apply various AI models that allow pharma teams to maintain their processes under control, react over possible deviations, and enhance the knowledge of their process.

At Bigfinite, we provide an integral solution that covers the essential tools for pharma to speed up the process development and manufacturing of vaccines. Once the data is properly consolidated in a centralized platform, ready to be used in GxP environments, a deployment sequence of AI practices brings the right knowledge to oversee the entire process. From the identification of relevant parameters to the generation of predictive insights, AI tools bring science and statistics to make sure each batch is right the first time.

This article was created in collaboration with the sponsoring company and the Xtalks editorial team.