In modern drug discovery, looking different is no longer enough.
MindWalk Holdings Corp. (NASDAQ: HYFT) says it has identified a growing and underappreciated risk facing pharmaceutical R&D: functional adjacency—situations where entirely different molecules deliver the same therapeutic effect, even when traditional sequence analysis suggests they are unrelated.
Using its proprietary HYFT® (Hyfte Functional Topology) technology, the company demonstrated how biological function can remain conserved even as genetic sequences diverge dramatically. In a recent influenza-focused application, MindWalk identified a stable biological “signature” that persists despite viral mutation, reinforcing the company’s thesis that function—not sequence—defines real-world competition, IP exposure, and asset value.
As AI-driven protein design accelerates, MindWalk argues this issue is becoming economically material—and increasingly visible in courtrooms, boardrooms, and deal models.
Why Functional Adjacency Is Becoming a Pharma Problem
For decades, drug discovery and IP strategy have leaned heavily on sequence alignment as a proxy for novelty, differentiation, and competitive distance. If molecules didn’t look alike on paper, they were often treated as distinct.
That assumption is breaking down.
Functional adjacency describes a scenario where different molecular structures converge on the same biological outcome—binding the same target, producing the same clinical effect, or competing for the same patient population. In practice, that convergence can erase perceived differentiation even when sequences appear novel.
According to MindWalk CEO Jennifer Bath, Ph.D., this convergence now represents a significant blind spot.
As generative AI tools enable teams to rapidly design thousands of candidate proteins, the likelihood increases that multiple “distinct” sequences end up solving the same biological problem in essentially the same way. When that happens, companies may unknowingly collide—commercially, legally, and strategically.
The Legal Backdrop: Courts Are Already There
Recent U.S. patent decisions have sharpened the stakes.
In Amgen v. Sanofi, the U.S. Supreme Court invalidated broad, functionally defined antibody claims targeting PCSK9 inhibition for lack of enablement. The ruling underscored a critical reality: claiming functional territory without adequately teaching how to make and use the full scope is increasingly untenable.
Similarly, in Juno Therapeutics v. Kite Pharma, the Federal Circuit invalidated certain CAR-T claims for insufficient written description, citing a lack of representative examples across the claimed functional range.
Taken together, these cases send a clear signal to the industry:
- Functional equivalence—not sequence similarity—often defines competitive overlap
- Sequence novelty alone is no longer a reliable shield
- IP strategies built on broad functional claims face growing scrutiny
- Asset valuations tied to perceived uniqueness may be more fragile than assumed
MindWalk positions functional adjacency as the common thread running through these outcomes—a risk that traditional bioinformatics tools were never designed to surface.
HYFT: Moving Beyond Sequence Thinking
HYFT technology is MindWalk’s answer to that gap.
Rather than comparing molecules based on linear sequence similarity, HYFT evaluates conserved functional patterns shaped by biophysical constraints, including structure, charge distribution, and binding interactions. These constraints can persist even when evolutionary relationships are unclear or sequences diverge dramatically.
In the influenza analysis, HYFT identified functional architecture that the virus appears to preserve to remain infectious—despite ongoing mutation. That observation supports MindWalk’s broader vaccine design strategy, aimed at identifying targets that offer broader, strain-agnostic protection.
More importantly, it illustrates the core value proposition: HYFT can detect meaningful biological similarity where traditional tools see none.
As MindWalk CTO Dirk Van Hyfte, MD, Ph.D., explained, the technology enables researchers to step outside sequence-centric thinking and focus on the functional constraints biology cannot easily abandon.
AI Is Making the Problem Worse—and More Urgent
Ironically, the same AI advances accelerating drug discovery are also amplifying the functional adjacency problem.
High-throughput generative platforms can rapidly explore functional space, producing diverse molecular sequences that converge on the same activity. From a design perspective, that’s a breakthrough. From a competitive and IP standpoint, it’s a minefield.
Without tools that detect functional convergence early, organizations risk:
- Discovering late-stage competitors they never saw coming
- Overestimating differentiation during licensing or M&A
- Filing patents vulnerable to enablement or written description challenges
- Allocating capital to overcrowded functional territory
MindWalk argues that functional intelligence must now sit alongside sequence, structure, and clinical data as a core input to R&D and corporate decision-making.
Beyond Influenza: A Platform, Not a Point Solution
The influenza findings build on MindWalk’s previously disclosed HYFT-enabled dengue epitope program, reinforcing the view that HYFT is a cross-pathogen platform rather than a single-asset discovery tool.
Across these programs, HYFT is positioned as a way to uncover conserved “design rules” in biology—patterns that persist because they are required for function, not because they share a common sequence origin.
That distinction matters for scale. If functional rules can be identified systematically, HYFT becomes applicable across therapeutic areas, modalities, and stages of development.
MindWalk frames the technology as a strategic intelligence layer within its broader biological reasoning and data platform, unifying sequence, structure, function, and literature into a single operational framework.
Strategic Implications for Pharma and Investors
MindWalk sees HYFT as relevant well beyond the lab bench.
The company says the technology is designed to help organizations:
- Identify functional competitors earlier, including AI-designed or convergently evolved assets missed by sequence analysis
- Strengthen in-licensing and M&A diligence, by assessing functional overlap rather than superficial novelty
- Inform patent strategy, especially under heightened scrutiny of function-based claims
- Improve portfolio and capital allocation decisions, by identifying crowded functional spaces before they become obvious
For investors, this reframes risk. Assets that appear differentiated on paper may, at a functional level, be far closer to competitors than expected—raising questions about long-term exclusivity and pricing power.
A Shift From Backward-Looking to Forward-Looking Analysis
At a higher level, MindWalk is advocating for a philosophical shift in how biopharma evaluates opportunity and risk.
Sequence alignment is inherently backward-looking—it asks what looks similar today. Functional intelligence is forward-looking—it asks what will behave similarly in the clinic, in the market, and under regulatory scrutiny.
As AI accelerates therapeutic design, that forward-looking lens may become essential.
MindWalk plans to engage pharmaceutical, biotechnology, and related organizations to explore collaborations and commercial arrangements that embed HYFT-based functional intelligence into discovery, diligence, and portfolio workflows.
The Bottom Line
AI has made it easier than ever to design new molecules. It has also made it easier than ever for those molecules to collide—functionally, legally, and commercially.
MindWalk’s HYFT technology is a bet that the next competitive advantage in drug discovery won’t come from making things look different, but from understanding when different things act the same.
In an era where functional convergence is accelerating and courts are paying closer attention, that insight may prove as valuable as any single drug candidate.
Power Tomorrow’s Intelligence — Build It with TechEdgeAI
