AI is no longer just a tool for analyzing results; it's becoming a scientific thinking partner that helps researchers design better experiments and interpret findings in real time. At Tech Week Grand Rapids, Ferris State University is showcasing how artificial intelligence can be woven directly into the design-build-test-learn cycle, a fundamental workflow in biotechnology research.

What Are Polyketide Synthases and Why Do They Matter?

Polyketide synthases, or PKS systems, are massive, assembly-line enzymes that manufacture chemicals essential to American biotechnology, including medicines, fuels, and bulk chemicals. Think of them as molecular factories: they take raw materials and assemble them into useful compounds. The challenge is that these enzymes are enormous and incredibly complex, making them difficult to engineer and improve.

Traditionally, scientists follow a repetitive cycle: design an experiment, build genetic constructs, test the results, and interpret what happened. Then they start over. This cycle can take weeks or months, especially when researchers must manually analyze data and decide what to try next. That's where AI enters the picture.

How Can AI Speed Up the Scientific Discovery Process?

Rather than replacing scientists, AI is being positioned as a scientific sounding board that accelerates each phase of the research cycle. The key insight is that AI doesn't sit above the scientific process; it sits within it, supporting hypothesis generation, experimental interpretation, and strategy selection.

Two student-led presentations at Tech Week Grand Rapids will demonstrate practical applications of this approach. The speakers will showcase how AI can assist with several critical tasks:

• Hypothesis Generation: AI helps researchers anticipate what might happen in an experiment before they run it, reducing wasted time on unlikely approaches.
• Experimental Interpretation: After running tests, AI helps scientists understand what the data means and why certain outcomes occurred.
• Strategy Identification: Machine learning identifies which engineering approaches are most promising based on patterns in existing data.
• Experiment Selection: AI recommends which experiments to run next, prioritizing high-impact tests over low-probability ones.

This approach, called "AI-in-the-DBTL-cycle," treats artificial intelligence as a research assistant rather than a replacement for human judgment. Undergraduates working on PKS engineering projects report that AI helps them anticipate experimental outcomes and interpret results more quickly, compressing what might take weeks into days.

Why This Matters Beyond the Lab

The implications extend far beyond academic research. Pharmaceutical companies and biotech firms face similar challenges when engineering proteins for drug manufacturing. If AI can help academic researchers move faster through the design-build-test-learn cycle, the same approach could accelerate drug discovery and development timelines in industry settings. This is particularly important for rare diseases and personalized medicine, where traditional development cycles are too slow and expensive.

The Ferris State sessions are part of a broader Tech Week Grand Rapids event running from September 14 to 19, featuring approximately 160 free, public events including panels, workshops, and demonstrations. The week explores how AI, data, and emerging technology are transforming clinical care, drug discovery, and patient outcomes across the greater Grand Rapids region.

Beyond protein engineering, other sessions will address how AI is being deployed in healthcare settings. One presentation will showcase the CHARM initiative, which stands for Collaboration to Harmonize Antimicrobial Registry Measures. This project demonstrates how AI can transform healthcare data into actionable insights using interactive dashboards and analytics, including locally hosted large language models that standardize prescription data while keeping sensitive information secure.

The broader message from Ferris State's Tech Week contributions is clear: AI's value in science and medicine isn't about replacing human expertise. Instead, it's about embedding machine learning into existing workflows so researchers can work smarter, faster, and with greater confidence in their decisions. For drug discovery specifically, this could mean shorter timelines from initial concept to clinical testing, potentially bringing life-saving treatments to patients years earlier than traditional methods allow.