In the high-stakes world of pharmaceutical development, the traditional "fail-fast" approach has long been hampered by a critical bottleneck: the timing of Absorption, Distribution, Metabolism, and Excretion (ADME) profiling. For decades, medicinal chemistry teams have relegated these essential pharmacokinetic assessments to the "lead optimization" phase—a stage reached only after significant time and capital have been invested in a specific lead series.
However, a new collaborative venture between Ginkgo Datapoints, Tangible Scientific, and Inductive Bio is poised to disrupt this paradigm. By launching "ADME-One," a high-throughput platform that integrates AI-driven projection with automated laboratory workflows, the consortium aims to move pharmacokinetic decision-making to the very beginning of the drug discovery process: Hit Identification.
The Strategic Pivot: Bringing PK to the Forefront
The fundamental philosophy behind ADME-One is to democratize access to high-fidelity pharmacokinetic data. Historically, the cost and logistical complexity of running comprehensive ADME assays on early-stage compounds meant that only a handful of "priority" candidates received the full diagnostic treatment.
"We asked ourselves: Could we pull together all the assays needed to get your first projection of human PK at a price point where you would now be doing this on most, if not all, of the compounds coming through?" explains Dr. Alex Taylor, Head of Medicinal Chemistry at Inductive Bio.
By integrating Ginkgo’s automated Tier 1 assays—covering microsomal stability, cell permeability, kinetic solubility, CYP inhibition, and plasma protein binding—with Tangible Scientific’s sophisticated compound management, the platform transforms ADME from a retrospective checkpoint into a prospective design tool. This shift allows medicinal chemists to gauge potential potency and human dose projections while the program is still in the ideation phase, rather than months later after costly synthesis cycles have already committed the team to a specific chemical scaffold.
A Chronology of Discovery: From Siloed Data to Integrated Workflow
The evolution of drug discovery has seen a transition from manual, bench-top synthesis to high-throughput screening, and now, to the age of "digital-first" medicinal chemistry. The emergence of the ADME-One consortium represents the next logical step in this timeline:
- The Traditional Era: ADME was an afterthought, performed only on lead candidates that survived initial potency screening. This frequently resulted in "late-stage attrition," where compounds were discarded after millions of dollars were spent due to unforeseen metabolic liabilities.
- The Automation Wave: The rise of robotics allowed for faster assay turnarounds, but the data remained fragmented. Researchers had to interpret disparate, raw readouts from different vendors, often lacking the context of how these metrics correlated to human clinical success.
- The Consortium Launch: With the introduction of ADME-One, the process is now centralized. Tangible Scientific manages the physical intake and logistics; Ginkgo Datapoints provides the high-throughput, automated experimental backbone; and Inductive Bio provides the "Compass" platform, which synthesizes these experimental results into a cohesive human PK projection.
- The Future Integration: As the consortium gathers more data, the machine learning models at the core of the platform become increasingly predictive, creating a "virtuous cycle" where every experiment informs the next, effectively raising the barrier for entry for poor-quality chemical matter.
Supporting Data: The Case for Early Dose Optimization
The emphasis on early-stage human dose projection is not merely a technical preference; it is a clinical necessity. Dr. Taylor notes that for experienced medicinal chemists, "dose is ultimately the thing you want to optimize for."
Clinical data strongly supports this focus. Research published in Hepatology and other leading journals has consistently linked high daily doses, particularly when combined with high lipophilicity, to an increased risk of drug-induced liver injury (DILI). The "rule-of-two" underscores that while high-dose drugs can be safe, they carry a significantly elevated risk profile compared to their lower-dose counterparts.
Furthermore, the operational benefits of low-dose, high-potency drugs are clear. Simpler dosing regimens are directly correlated with higher patient adherence rates—a critical factor in the success of any chronic therapy. Despite this, the industry has historically struggled to integrate these complex, multidimensional data points early enough to influence compound selection. ADME-One bridges this gap by providing a quantitative basis for these decisions, allowing teams to distinguish between compounds that are simply "potent" and those that are "drug-like."
Navigating the Complexity of Metabolic Profiles
One of the most persistent challenges in medicinal chemistry is the "scorecard fallacy"—the tendency to reject compounds that perform poorly in a single assay without considering the holistic profile.
The history of triazole antifungals serves as a perfect case study. Fluconazole and Itraconazole represent opposite ends of the pharmacokinetic spectrum. Fluconazole is polar, clears quickly through the kidneys, and exhibits low protein binding. Itraconazole, conversely, is highly lipophilic and heavily protein-bound. If judged solely by standard, rigid criteria, both might have been discarded at the hit ID stage. Yet, both became cornerstone therapies.
The ADME-One platform aims to prevent this type of premature rejection by providing a more nuanced, systemic view of a molecule’s properties. By analyzing the balance of all pharmacokinetic parameters, the platform helps researchers identify compounds that, while not "perfect" in every single assay, possess the ideal composite profile for therapeutic success.
Official Responses and Industry Implications
The launch of ADME-One arrives at a time of significant economic and geopolitical shift in the life sciences sector. The pressure to remain lean and cost-conscious is universal, but it is now compounded by the need for data sovereignty.
The Onshoring Mandate
As U.S. and European developers move preclinical operations back onshore—partially in response to the BIOSECURE Act and the growing demand for secure supply chains—the ADME-One platform offers a distinct competitive advantage. By running the entire workflow in the U.S. and delivering results in days rather than the weeks typically required by offshore Contract Research Organizations (CROs), the consortium directly addresses the logistical and security concerns of modern drug developers.
The Consortium Security Model
A primary concern for any company contributing data to a shared model is intellectual property security. Dr. Taylor explains that Inductive Bio has addressed this through a sophisticated legal and technical framework. The consortium operates under a strict data-pooling agreement where no partner can see the proprietary chemistry of another.
"We put substantial engineering into making the pooled data impossible to reverse-engineer," Taylor states. By training global models on pooled data and then fine-tuning "local" models for each individual client, Inductive Bio ensures that clients benefit from the collective intelligence of the consortium without sacrificing their unique competitive edge.
Conclusion: The Virtuous Cycle of Discovery
The ultimate vision of the ADME-One consortium is to transform the way medicinal chemistry is practiced. By moving the most critical pharmacokinetic questions to the beginning of the funnel, the platform reduces the "sunk cost" bias that often traps teams in unproductive research paths.
"Drug discovery is science at the end of the day, and science is not engineering," says Dr. Taylor. "You can give your best guess of what a compound is going to do, but at the end of the day you need to reduce it to practice, synthesize it, and test it."
In this sense, AI and automation do not replace the scientist; they empower the scientist to make better, more informed choices about which compounds deserve to be synthesized. In an era where every dollar spent in the lab must be justified by its potential for clinical success, the ability to predict, measure, and optimize for human PK at the hit identification stage is not just an efficiency gain—it is the new standard for modern, responsible drug discovery. As the ADME-One platform continues to ingest data, its predictive power will only grow, potentially ushering in a more efficient, data-driven era for the entire pharmaceutical industry.
About the Author
