Whitepaper

In-silico bioprocess modeling for CHO process development

Process development forces high-impact decisions from limited data — which media, feed strategies, inoculum conditions and set points are most likely to improve yield and robustness. Brute-force screening can't cover the space: even a moderate number of media variants, feed settings and inoculation strategies produces a design that's too large, too slow, and too expensive to run.

This paper describes a modeling framework that learns a compact representation of bioprocess dynamics from historical run data, then acts as a digital twin — simulating expected trajectories from planned set points alone. Trained on historical fed-batch CHO runs spanning 5–500 L, it lets teams evaluate candidate conditions in silico and take only the most informative ones to the bench.

Get the whitepaper

Fill this form to receive the whitepaper:

Thanks — it's on the way.

Check your inbox for the whitepaper. If it doesn't arrive in a few minutes, look in your spam folder.

Oops! Something went wrong while submitting the form.

Speaker

Yaron David, Co-Founder and CTO of BioRaptor

Yaron David, MD, PhD,

Co-Founder and CTO, BioRaptor

You’ll  learn:

<ul><li>Why purely empirical screening and purely data-driven models both break down in small, unevenly distributed process datasets</li><li>How joint representation and dynamics learning captures process state as it evolves, not just endpoints</li><li>How the model simulates VCD, metabolites, volume and product concentration from set points alone</li><li>How to use predicted trajectories and confidence bands to prioritize a DoE and avoid low-value runs</li><li>Where the model flags uncertain regions that warrant more experimental data</li></ul>

@ 2025 BioRaptor. All rights reserved

Privacy policy 
Linkedin logo