- How it Works
- Applications Overview
- Bioprocess Development
- Molecular Biology
Our Design of Experiments (DOE) Masterclass series has been a huge success! Thank you for registering and attending. If you missed it, you can watch the recording here.
We got a lot of questions during the webinar. The most popular was along the lines of “How do I get started?” Because we’ve only got so much time, we couldn’t answer it in detail during the webinar but we can do so in this blog entry.
The best way to get started using DOE is simply… to start. As with all things, the journey of a thousand miles begins with the first step. But what’s the best first step?
Ask These Five Questions Before Starting
We begin by thinking critically about the system or process we want to study. These are the five most important questions we can begin with:
- What is the response that we are trying to measure or learn more about?
- What are the factors that can affect these responses that we want to investigate?
- What are the constraints on our experiment?
- Are there factors that we can’t change or resources that are limited?
- How many experimental runs can we perform?
These five questions will inform which type of design is best suited for what we want to do. Coming up with a design for a DOE is never a walk in the park, but thinking about and addressing these questions beforehand will make everything run a lot smoother.
There are a number of statistical software options that can simplify the complexity inherent in DOE and help set up an experimental run matrix. Software tools like JMP, Minitab, or Design-Expert by Stat-Ease have excellent features for generating powerful multifactorial statistical designs to help map out which experimental runs to execute in the lab.
Take Inspiration from Others
Exploring case studies of how others have leveraged DOE in the physical/experimental setting can be a great way to understand practical applications of statistical design. These case studies can also help you understand the analytical benefits of one design type over another.
Learning about the best ways to leverage a particular statistical design space and how to determine factors and levels for investigations that meet your experimental needs or limitations is what makes the DOE journey equal parts exploration and optimization. As your understanding of running and analyzing a DOE improves, so too will your ability to examine a system, understand the constraints, and apply the appropriate design types.
Get Into the Literature
Perhaps one of the best ways to get started with Design of Experiments is finding the best books on the topic.
Practical Design of Experiments: DOE Made Easy by Colin Hardwick is an excellent resource if you are new to DOE. It will show you how DOE is used to tackle some of the most challenging problems in science and engineering that are out there right now. This book provides a generic yet comprehensive approach to DOE, making it useful in any scientific domain.
Another good book is Optimal Design of Experiments: A Case Study Approach by Peter Goos and Bradley Jones. As the title suggests, it takes a practical approach. Each chapter reads like a short story, documenting a situation and how a custom-designed experiment solved the problem. The second half of the chapter then illustrates the statistical details involved.
Getting Started with DOE
DOE is part of Synthace’s core and history. We’ve long championed it as one of the best approaches to unlocking the complexity and mystery in life sciences. But we also recognize that it isn’t always an easy way to work, especially when multiple automated instruments and analytical devices are involved. This is one of the reasons we don’t see it applied as robustly in the life sciences industry as others.
The Synthace platform offers biologists and engineers a way to streamline the execution of DOE on multiple devices without needing to hard-code devices. It also automatically structures the resulting data with the experimental design and metadata in an analysis ready format. Using our software helps speed up the execution of DOE and cuts down on error-prone sample and data handling.
We continue to advocate for scientists and engineers to take advantage of DOE in their scientific field of study. Be sure to check out our DOE Masterclass and reach out if you have any questions about how it can transform your work and take your R&D to the next level.
Tag(s): Design of Experiments (DOE)
Other posts you might be interested inView All Posts
Design of Experiments (DOE)
Here’s Why You Should Stop Doing Science Like You Were Taught in High School (Part 1)
Synthace Expert Interviews: Cell and Gene Therapies with Dr Damian Marshall
Earth Day 2022: We Need Biology in the Fight Against Climate Change. But Are We Ready To Use It?