Design for Manufacturing: A New Paradigm for Computational Life Sciences

Biotechnology is going through a paradigm shift and is evolving from its inception of the 1980s. We are no longer beholden to discovering the best that nature has to offer and today we can push biological systems to do things that evolution never imagined. Much like how the industrial revolution brought us scaled manufacturing and synthetic chemistry, biotechnology now offers us access to products and medicines previously unimaginable. But we are just now in the early stages of this new industrial maturity cycle, where we no longer need to ask “is it possible” and we can now ask “how can we do this at scale”. 

‍

As traditional industrial manufacturing went through this maturity cycle over the past century, the paradigm of design for manufacturing emerged to enable higher throughput, lower cost, and faster delivery of products across the globe. When standardization emerged as the common principle of manufacturing - engineering multiple parts and products to fit into the same manufacturing lines - we achieved an explosion of progress and products.

Biotechnology, however, is just reaching this point in its maturity cycle, where advances in data capture, computational tools, and manufacturing science can create standardization across products and industries. This new paradigm for the biotechnology industry will be driven in large part by computational life sciences, and building manufacturing-based thinking earlier into the discovery process that will directly improve lives. ‍

Discovery

We have only started to scratch the surface of what we can do to tackle disease, increase human longevity, and create new products by means of genetic engineering. Up to this point, most of scientific discovery has been focused on improving the efficiency of a new product, whether that means efficacy of a new drug or production of a new product, and this focus leads to manufacturability being a factor only after an initial demonstration of success at the bench top. But this design is often done at very small scales at the lab bench where a few milliliters of solution carries the experiments on which we base progress. Unlike traditional manufacturing, however, small scale experiments rarely scale easily from bench top to industrial production because we are working with living systems whose thermodynamics, nutrient requirements, and life cycles are drastically affected by their environment.  ‍

Manufacturing‍

Taking a product from bench to manufacturing requires a painstaking process of transferring the protocols to a production facility and designing the systems to scale up production and eventually purify the product via downstream processing. Many products require bespoke processes where the development costs millions of dollars per product. Unfortunately, it's not uncommon for a product to fail at this manufacturing stage because the biological system doesn’t scale the way we intend it to. This leads to long and expensive product development cycles that delays time to market which directly impacts the lives of those who may depend on that product. 

Design for Manufacturing

The paradigm of design for manufacturing, which is the practice of designing products with the specific aim to improve manufacturability, is well known in traditional manufacturing. This reduces the time to market and cost to manufacture a product. To do this in the life sciences, we need to start building manufacturing considerations into our design and discovery phases. 

Unlike traditional manufacturing, biotechnology deals with living systems where trade offs can make or break a product's viability. But in many cases, such as when optimizing the use of codons in a genetic construct, small changes can alter a product's ability to be manufactured while not changing the final product. 

Let's consider an example. A company producing gene therapies can make various design considerations when building their constructs. These changes at the base pair level can affect how a constructed secondary structure affects its production efficiency, how its genetic content affects immune response, and how optimizing across these factors can yield higher production of the same end product because variations in genetic elements can still code for the same protein structures. One construct just might be better at doing so at scale than another. 

All these factors are not easy to efficiently optimize across. There can be thousands of variations to consider and the order of how you optimize might affect the end result. In this case, computational tools can test thousands to millions of hypotheses simultaneously while optimizing different aspects of the product in parallel. All before a product goes to manufacturing. Computational life sciences can start to build manufacturing considerations earlier into the process for all our product development efforts.

The sooner we can build these paradigms into our design and discovery processes, the more likely we are to avoid costly manufacturing failures and the faster we can get products to market. 

In an industry where products drastically affect our health and well being, new paradigms can be a life saver. 

‍

Interested in modernizing your computational abilities? Schedule a demo with our team.