Discover the opportunities with world-leading data analysis
Better characterize raw materials to minimize variability
Improve fermentation for higher yields with experimental design
Increase product quality with real-time process and quality control
World-leading data analysis software for bioscience companies
For decades, we have provided tools to biotechnology companies and research institutes around the world to bring their data to life.
Our advanced multivariate data analysis software, The Unscrambler® X, enables researchers and engineers to identify underlying patterns and analyze complex data, quickly and easily. This lets them address challenges such as understanding raw material variations or controlling the complexities of the fermentation process.
Multivariate data analysis and design of experiments are used for a wide range of applications in the life sciences, food & beverage, chemicals, pulp & paper, agricultural, automotive and energy sectors, enabling companies to reduce R&D timeframes, optimize production processes and improve quality control. With your leading edge science and our advanced data analysis software, we can help you drive innovation and get your leading biotech products to market faster.
“We use the Unscrambler to make sense of the large data sets we generate from very complex samples”
Dr Garth Maker, Lecturer in Pharmaceutical Chemistry, Separation Science and Metabolomics Laboratory, Murdoch University, Australia
Multivariate data analysis and design of experiments can be used in most biotechnology fields.
- Characterize raw materials using classification and prediction models
- Manage scale up from pilot plant to production with QbD principles
- Analyze data for biomarker discovery and diagnostics in medicine
Bio-chemicals and Bio-plastics
- Improve quality with spectroscopic instruments & multivariate calibration
- On-line process monitoring enabling timely adjustments to fermentations
- Develop renewable bio-chemicals more efficiently with designed experiments
Foods and Agriculture
- Determine the presence of pests or fungi using instrumental techniques
- Identify the relationships between local conditions, crop genetics and yield
- Develop more nutritional foods more efficiently with design of experiments
- Create high value blending stock through better process understanding
- Real-time classification of feedstock grade for faster refinery acceptance
- Detect impurities such as methanol with multivariate calibration
EXAMPLE APPLICATIONS OF MULTIVARIATE ANALYSIS
Increasing bio-fuel refinery efficiency
To operate efficiently, bio-fuel refineries require raw material of a consistent quality. Near Infrared (NIR) spectroscopy results are analyzed with multivariate exploratory data analysis and regression analysis to determine the quality of raw materials based on calibration samples. Additionally, refineries can use multivariate classification models to grade raw materials into clusters of high, medium or low grade, enabling the refinery to Pass or Fail the raw material immediately upon receiving stock, further improving operational efficiency.
Identifying metabolites in bio-medicine
Multivariate methods are increasingly being used for biomarker discovery and diagnostics in medicine, such as metabolites that may indicate the presence of cancer cells. Exploratory data analysis methods such as Cluster Analysis and Principal Component Analysis (PCA) as well as classification methods such as Support Vector Machine Classification (SVM-C) are used for pattern recognition to identify which metabolites are contributing to variance, highlighting which biochemical pathways are involved.
Optimizing crop yields in agricultural production
Multivariate data analysis can be used to better understand the nutritional content of crops, thereby increasing the crop yields and ensuring the global food supply. Soil spectroscopy is heavily reliant on quantification with regression methods such as Partial Least Squares Regression (PLSR) and Support Vector Machines. This allows researchers to predict the vital chemical elements that indicate whether land is suitable for agricultural production and the specific crops ideal for the conditions.
Process monitoring in bio-pharmaceuticals
The combination of design of experiments (DoE) and multivariate analysis (MVA) gives detailed insight into the processes by analysis of historical data followed by experimental plans to verify the hypotheses in lab or pilot scale. Validation across raw material batches can be done using categorical variables which shows whether a production process is robust towards raw material variations. One specific application is in the production of a pharmaceutical product. Together with an understanding of the underlying chemistry, DoE gives quantitative information regarding the process settings for improving yield in the synthesis.
What is multivariate data analysis?
Biotechnology is an extremely complex science involving a diverse range of micro-organisms and enzymes with a potentially limitless number of applications. Because biotechnology systems are typically comprised of a large number of related variables, looking at each variable in isolation does not usually show the full picture. This is why multivariate data analysis is an essential tool for biotechnology companies.
Multivariate analysis is the investigation of many variables simultaneously, in order to understand the relationships that exist between them. While traditional (univariate) statistical approaches serve their purposes for investigating and understanding simple systems, when the relationships between variables are complex, as in biotechnology, a single variable cannot adequately describe the system.
Exploratory data analysis (data mining), clustering, regression and predictive analysis are typical multivariate tools which help biotechnology researchers and engineers better identify and understand the complex relationships and variables in their data.
What the experts say:
Our powerful multivariate data analysis software enables scientific and research organizations to understand their complex data faster and easier. But don’t just take our word for it, read the reviews from these leading scientific titles.
“This intelligent engine borders upon data mining, as it cuts through prediction and classification problems”
Scientific Computing 2011
“Cutting through complex data sets to underlying structures…is simplicity itself”
Scientific Computing 2011