Collaborations at AnimalScienceDataLab

COLLABORATIONS

We collaborate with the livestock industry to develop data-driven solutions for animal health, nutrition, and production challenges. Whether you need support analyzing data, identifying hidden patterns, or developing interactive decision-support tools, we can help.

READY TO GET MORE FROM YOUR DATA?

Fill out the form below to request a free consultation to explore how data-driven solutions can help you make better decisions in animal health, nutrition, and overall production performance.

NameEmail AddressInsitution / CompanyPosition / RoleWhat support are you looking for?Statistical AnalysisMachine Learning / AIInteractive Software & Analytical ToolsData MiningOtherComments

Thank you! Your submission has been received!

Oops! Something went wrong while submitting the form.

FREQUENTLY ASKED QUESTIONS (FAQs)

What type of collaborations can companies develop with Animal Science DataLab?

Animal Science Data Lab is a platform that helps identify potential collaborations between livestock industry partners and the University of Minnesota (UMN). The goal is to start conversations with companies, understand their questions, evaluate the type of data available, and determine whether a project can be developed through the appropriate UMN process.
‍
Depending on the question, the data available, and the expected outputs, collaborations may involve data analysis, machine learning, mathematical modeling, decision-support tools, controlled animal research trials, validation studies, or broader research partnerships.
‍
The following list provides examples of UMN collaboration pathways that may apply depending on the type of project, the data involved, the expected outputs, and the level of engagement between the industry partner and UMN. We work closely with collaborators to evaluate data, answer research or production questions, and help apply results to specific problems.

Industry-sponsored research: A company funds a research project at UMN involving data analysis, modeling, validation studies, animal studies, or decision-support tools.
Research collaboration: UMN and an industry partner work together by contributing expertise, data, research questions, and interpretation.
Service agreement: A company requests a defined service, such as data cleaning, statistical analysis, modeling, reporting or dashboard development.
MN-IP sponsored research: A company sponsors research that may generate software, models, tools, methods, or other intellectual property.
Master research agreement: A broader agreement that supports multiple projects with the same company over time.
Controlled animal research trials: A company evaluates a product, feed additive, nutritional strategy, vaccine, diagnostic tool, health intervention, or management practice in animals.
Corporate engagement: Companies connect with UMN researchers to identify shared interests and explore possible collaborations.
Technology licensing or commercialization: A project leads to software, models, tools, or technologies that may be licensed or commercialized through UMN processes.

All collaborations would be developed through the appropriate UMN process based on the type of project, funding structure, data involved, and required approvals.

What agreements or documents may be needed to start a collaboration?

Before a project begins, some collaborations may require supporting agreements or project documents. These documents help define how data, confidential information, materials, intellectual property, reporting, and compliance requirements will be handled.

Before promising specific results, we also perform an initial data quality assessment to evaluate whether the available data have enough structure, consistency, and meaningful variation to support the proposed analysis. If the project appears feasible, Sponsored Projects Administration (SPA) or other relevant UMN offices can help develop a tailored agreement or contract based on the type of collaboration and the needs of the project.

Possible supporting agreements or project documents include:

Data Use Agreement: Used when company data are shared with UMN for research or analysis. It can define how data are shared, protected, accessed, analyzed, reported, and potentially published.
Confidentiality Agreement/Nondisclosure Agreement: Used when confidential information needs to be discussed before or during a project. It can protect company information, unpublished data, proprietary ideas, or sensitive project details.
Intellectual property terms: Used when a project may generate software, models, methods, tools, inventions, or other intellectual property.
Publication and reporting terms: Used to clarify how results may be reported, reviewed, shared, or published.
Compliance and approval documents: Used when a project requires animal care approval, biosafety review, data security review, data security review, or other UMN compliance processes.

All agreements and documents would be developed through the appropriate UMN process based on the type of project, the data involved, the expected outputs, and the required approvals.

What process do you follow when working with collaborators?

The collaboration process starts with a conversation and then moves step by step toward a formal project through the appropriate University of Minnesota (UMN) process.

The goal is to make sure the project is clearly defined, the available data can support the proposed analysis, and the expected outputs are realistic and useful for the industry partner. We work closely with collaborators to evaluate data, answer research or production questions, and help apply results to specific problems.

Below are the steps we follow when working with collaborators:

Free Consultation

We discuss your goals, questions, and available data.

Data Quality Assessment

We evaluate whether the data supports meaningful analysis.

Analysis & Modeling

We apply statistics, AI, and/or machine learning approaches.

Interpretation

We translate results into biologically meaningful findings.

Decision Support

We help convert results into tools, reports, or strategies for better decisions.

How do you assess whether my data is suitable for AI analysis?

We collaborate with the livestock industry to apply AI and machine learning methods to identify hidden patterns in biological and production data and support research and production decisions. However, the first step is always to assess data quality. We evaluate whether the dataset has enough structure, consistency, and meaningful variation to support AI analysis.

The video below provides an overview of how we assess data quality in our projects

What types of statistical analyses and modeling methods do you apply?

We collaborate with the livestock industry to develop statistical and mathematical models that help explain biological systems, evaluate production strategies, and support data-driven decision-making. Our work includes growth modeling, predictive analytics, time-series analysis, and customized analytical workflows for research and production applications.

Below are some examples of how these approaches have been applied in collaborative projects:

In collaboration with Schwartz Farms, we applied time-series analysis to commercial production data to evaluate the impact of Mycoplasma hyopneumoniae infection on swine production sustainability and performance metrics. Publication DOI: 10.1093/jas/skaf300.290.
In collaboration with Turner Institute of Ecoagriculture and South Dakota State University, we analyzed more than 20 years of commercial bison production data to study how climate and translocations influence growth and body size in North American bison. Publication DOI: 10.3389/fevo.2025.1683267.

How can interactive tools support research and production?

We develop interactive tools that help researchers, students, and producers work with data, evaluate different scenarios, and apply results through simple and customizable interfaces. These tools can support activities such as nutrient evaluation, production planning, diet formulation, and sustainability analysis.

Below are some examples of tools developed through collaborative projects:

In collaboration with PIC, we developed a sow feeding and nutrient specification web application for highly prolific sows. [Publication DOI: 10.1093/jas/skab054.246], [App Link.]
Interactive diet formulation and life cycle assessment (LCA) tools developed for teaching in animal nutrition and sustainability analysis [App Link].