A novel self-learning, small-scale, high-throughput bioprocess platform for fast and optimal design of bioprocesses
Producing drugs/vaccines at scale requires manufacturing processes to be highly reliable & economically efficient. Yet process development is cost & time intensive, as a plethora of engineering possibilities require experimental investigation and optimization.
This project seeks to establish a self-learning, small-scale, high-throughput bioprocess platform for fast and optimal design of bioprocesses. It is based on m2p’s miniaturized high-throughput bioreactor device, the BioLector® (BL), and DataHow’s suite of unique knowledge centred machine-learning algorithms, which are being fully integrated.
High-throughput experimental Data can be transferred from the BioLector® (BL) to DataHow’s software suite, where they can be directly analyzed and the processes modeled. Based on this insight, the next round of BioLector® (BL) experiments can then be automatically designed such as to achieve the desired criteria.
The self-learning platform that we seek to build is the first of its kind & hence highly competitive to the existing offering. The platform can change how early biopharmaceutical process development is done, significantly lowering costs by millions € per drug.