With AlphaFold, researchers are charting a new frontier in drug design, enabling more effective therapies than ever before. The insights come from leaders who bridge science and medicine, turning bench discoveries into bedside realities. One such leader is Karen Akinsanya, who serves as President of R&D, Therapeutics, at Schrödinger in New York City. She shares the formative AlphaFold story that frames how advanced computational biology is reshaping the development of therapeutic molecules. Her perspective reflects a driving commitment: move from laboratory breakthroughs to real patient benefits while confronting the persistent burden of disease. The journey, she explains, begins with fundamental curiosity about biology and evolves into actionable design strategies that can change lives. This narrative ties together scientific innovation, patient-centered goals, and the practical challenges of bringing novel medicines to those who need them most. It is a story about the convergence of technology, biology, and clinical impact — a convergence made possible by AlphaFold and complementary, physics-based modeling approaches. Below, the themes of her experience illuminate how the field is advancing toward safer, more selective drugs that better address cancer, heart disease, and related conditions, while acknowledging the ongoing urgency this work seeks to meet in homes and hospitals around the world.
A personal journey at the crossroads of academia, discovery, and patient care
Karen Akinsanya’s professional trajectory spans both theoretical exploration and practical application. She emphasizes her dual exposure to academia and the drug discovery pipeline, noting that this blend has given her a unique vantage point. On one hand, studying proteins and genes provides a deep understanding of how disease-causing targets can be mitigated with therapeutic molecules. On the other hand, experience at the patient bedside reveals the real-world consequences when treatments reach those in need. She describes this as a continuous loop: laboratory insights inform clinical strategies, and patient outcomes, in turn, refine research directions. The overarching question she centers on is not merely how to design an effective molecule but how to improve the entire path from discovery to patient benefit.
In her view, cancer and heart disease remain persistent, daily challenges. Despite scientific advances, lives continue to be lost while researchers strive to translate discoveries into cures. This grim reality reinforces the imperative to optimize every step of the process. It is not enough to identify a promising target; the next critical task is to develop a molecule that can engage that target with high precision while minimizing unintended effects on other biological pathways. Akinsanya stresses the importance of efficiency, accuracy, and safety in drug design, acknowledging that improvements at each stage can accumulate into meaningful clinical gains. She often reflects on nature’s efficiency as a guiding principle: when a target is identified, the biological landscape around it often reveals related targets that resemble one another in structure or function. These relatives, while offering opportunities for therapeutic intervention, also pose the risk of off-target interactions that can complicate development and patient safety.
In this context, the challenge for drug discovery specialists is to craft molecules that selectively bind and modulate a specific member of a protein family—the receptor—that is most closely linked to the disease process, without inhibiting other family members. This selectivity is critical because while related proteins may share structural features, they can have distinct roles in physiology. A misstep in selectivity can lead to diminished efficacy or, worse, unintended side effects. Akinsanya highlights that achieving this balance requires a nuanced understanding of protein structures, dynamics, and interactions within the cellular environment. It is within this demanding landscape that AlphaFold has shown particular promise, offering a powerful way to refine our predictions about how targets behave and how potential drugs can influence that behavior.
Throughout her discussion, she returns to a central theme: the need to accelerate progress without compromising safety or scientific integrity. The imperative to shorten the time from discovery to patient benefit sits at the heart of the work at Schrödinger and similar organizations. This involves integrating advanced computational tools with traditional experimental validation, ensuring that predictions translate into tangible therapeutic advances. By combining cutting-edge software with rigorous scientific methodology, teams can rapidly iterate on design concepts, evaluate potential liabilities, and advance candidates with greater confidence. In short, her perspective frames AlphaFold not simply as a predictive aid, but as a strategic driver of more informed, efficient, and patient-centered drug development.
AlphaFold’s role in redefining modern drug discovery
AlphaFold has emerged as a transformative tool in the drug discovery toolkit, reshaping how scientists approach the structure and behavior of biological targets. Akinsanya points out that AlphaFold’s predictive capabilities, when integrated with established physics-based software, extend beyond static snapshots of single protein targets. They enable researchers to simulate not only the actions of individual family members but to explore how entire protein families interact and respond under different conditions. This holistic view helps researchers anticipate how a drug might alter a receptor’s activity across related proteins, thereby informing strategies to maximize therapeutic benefit while minimizing adverse effects. In this sense, AlphaFold contributes to a broader, more nuanced understanding of target biology.
From a practical standpoint, AlphaFold accelerates the early phases of design by providing high-confidence structural models that guide medicinal chemistry decisions. When researchers know the precise geometry of a binding pocket, they can more effectively tailor chemical structures to fit that pocket, optimizing interactions and improving binding affinity and selectivity. Such structural insights help reduce the number of iterative cycles required in the lab, saving time and resources. Yet Akinsanya emphasizes that AlphaFold is most impactful when used alongside physics-based simulations that capture the dynamic nature of molecules and their interactions. Proteins are not rigid; they move, adapt, and respond to the cellular environment. By combining AlphaFold’s structural predictions with atomistic simulations that account for thermal motion, solvent effects, and other forces, researchers can gain a more faithful picture of how a potential drug will perform in real biological systems.
One of the most compelling conceptual advantages of this integrated approach is the ability to move beyond a single-target paradigm toward a more systemic view of family behavior. In nature, receptor proteins often exist as families with shared features and diverse roles. A drug designed to engage a specific member must do so selectively, avoiding unintended engagement with relatives that could cause side effects or reduce efficacy. The AlphaFold-enabled clarity about structural similarities and differences across family members helps researchers map these relationships with greater confidence. The resulting design strategies can target subtle distinctions in binding site geometry, electrostatics, or dynamic behavior that differentiate one member from another. This level of discrimination is precisely what makes a drug not only potent but also safer and more selective in its action.
Akinsanya’s narrative illustrates how the scientific landscape has evolved with AI-driven structure prediction at its core. The combination of AlphaFold’s predictive power with robust, physics-based modeling creates a synergistic workflow. It enables the assessment of how alterations to a drug’s chemical structure may influence not just one target but an entire receptor family ecosystem. This approach improves the probability that a therapeutic will achieve its intended effect while minimizing collateral interactions that could undermine patient outcomes. In an era where precision medicine is increasingly prioritized, such capabilities are invaluable for delivering targeted therapies that align with individual disease biology and patient needs.
In practice, the AlphaFold-enabled paradigm supports several critical workflow improvements. It informs target prioritization by providing structural rationales for why certain receptors within a family are more druggable or more likely to yield a favorable therapeutic window. It enhances lead optimization by allowing designers to envision how changes in binding mode could influence selectivity across related proteins. It also supports risk assessment by highlighting potential off-target liabilities early in the design process, guiding preclinical strategies to mitigate those risks. Collectively, these benefits contribute to a more efficient and informed development path, enabling teams to progress from concept to candidate with greater assurance.
Target selectivity: navigating the receptor family landscape
The concept of a “target” in drug discovery is nuanced. Each therapeutic target often belongs to a larger family of related proteins, each member sharing structural motifs yet possessing distinct biological roles. The practical implication is clear: a drug that binds one member might, if not carefully designed, also interact with others in the same family. Off-target interactions can blur therapeutic effects, cause unintended physiological changes, or generate safety concerns. Akinsanya underscores that the central challenge for drug designers is to identify a compound that binds one family member effectively and specifically, while avoiding unnecessary engagement with its relatives.
This is where AlphaFold’s value becomes particularly pronounced. By delivering accurate, high-resolution models of receptor structures across family members, AlphaFold helps researchers detect subtle structural variations that distinguish one protein from another. These differences can be exploited to craft selective interactions, such as exploiting a unique pocket geometry, a distinct orientation of amino acid side chains, or a differential conformational flexibility that makes one receptor more amenable to binding than others. In addition, the ability to simulate how different family members respond to a given molecule provides insights into differential activity and potential selectivity liabilities. Such information is critical for prioritizing compounds and for guiding medicinal chemistry toward designs that optimize therapeutic index.
Akinsanya’s emphasis on the family-based perspective aligns with a broader strategic shift in drug discovery toward precision pharmacology. Rather than pursuing broad, non-selective engagement, modern strategies favor targeted modulation of disease-relevant pathways with careful consideration of context. The receptor family approach embraces this reality by recognizing that nuanced differences among family members can be leveraged to achieve desired outcomes. AlphaFold, when integrated with physics-based methods, becomes a powerful enabler of this nuanced, selective design. It supplies the structural cues needed to distinguish targets at a level of detail that would be difficult to obtain through traditional methods alone.
Moreover, the integration of these technologies helps researchers anticipate and mitigate risks early. If a proposed compound is predicted to fit several family members with high affinity, medicinal chemists can preemptively adjust the chemical scaffold to reduce off-target binding. Conversely, if a compound shows strong selectivity for a single member in silico, it can be prioritized for further development with greater confidence. In this way, AlphaFold-guided selectivity assessments contribute to a more predictable and streamlined development process, reducing late-stage bottlenecks associated with safety and efficacy concerns.
The synergy between AlphaFold and physics-based simulations
A core theme in Akinsanya’s narrative is the complementarity between AlphaFold’s structural predictions and physics-based software that models atomic interactions. This partnership is more than a simple summation of tools; it represents a holistic approach to understanding how drugs interact with their targets in a dynamic, biophysical context. While AlphaFold provides static, high-confidence structural blueprints, physics-based simulations add depth by revealing how those structures behave under physiological conditions, how binding pockets accommodate ligands, and how small changes in molecular structure influence binding energetics.
This synergistic workflow enables researchers to address questions that neither method could answer alone. For example, static models can reveal potential binding modes, but only dynamic simulations can show whether a ligand remains bound under physiological temperatures, how solvent effects modulate interactions, and how conformational changes in the receptor might alter binding affinity over time. The combined approach fosters more accurate predictions of potency, selectivity, and residence time—key determinants of therapeutic success. It also supports the exploration of structure-activity relationships in a more nuanced way, allowing researchers to iterate designs with a clearer map of how structural features translate into biological effects.
Akinsanya notes that the combined capabilities of AlphaFold and physics-based tools help scientists simulate not just single proteins, but how entire family groups behave within a system. This systemic perspective is particularly valuable when considering receptor families where crosstalk, redundancy, and compensatory mechanisms can influence drug performance. By understanding how different family members respond in concert, researchers can design interventions that achieve the intended therapeutic effect while minimizing compensatory pathways that might undermine efficacy. In practical terms, this means more informed lead optimization, better risk management, and a clearer path to selecting the most promising candidates for preclinical studies.
In addition, the integrated approach supports broader questions about drug design strategy. It helps identify structural or energetic features that confer selectivity, guides the choice of chemical scaffolds, and informs the design of functional assays that can validate computational predictions. It also fosters a more iterative dialogue between computation and experimentation, where predictions are tested, refined, and re-tested in a continuous loop. The result is a development workflow that is faster, more rigorous, and better aligned with the ultimate goal of delivering safe and effective medicines to patients.
From bench to bedside: translating computational insights into patient impact
The ultimate aim of this research paradigm is to shorten the path from discovery to patient benefit. Akinsanya’s perspective centers on the human dimension of drug development: the patients waiting for new, better therapies and the clinicians who depend on robust, reliable treatments that can withstand the test of real-world use. By enabling more precise targeting and reducing off-target risks, AlphaFold-enabled workflows contribute to drugs that are not only effective but also safer for diverse patient populations. This emphasis on patient outcomes, alongside scientific advancement, reflects a holistic view of what it means to translate computation into care.
The bench-to-bedside journey remains complex. It requires rigorous validation across multiple stages, including biochemical assays, cellular models, animal studies, and, ultimately, clinical trials. The predictive gains from AlphaFold and physics-based simulations must be substantiated by experimental data, ensuring that computational confidence translates into real-world efficacy and safety. Akinsanya’s experience underscores the importance of maintaining high standards of scientific integrity and iterative testing throughout this process. It is through this disciplined approach that the promise of computationally guided drug design becomes tangible improvement in quality of life for patients.
From the perspective of leadership and organizational strategy, integrating these technologies also involves cultivating interdisciplinary collaboration. Bringing together structural biology, medicinal chemistry, pharmacology, computational science, and clinical insights creates a richer decision-making environment. In such teams, AlphaFold models become shared references that anchor discussions about target viability, design directions, and risk management. The result is a more cohesive, informed, and agile development pathway that can respond to new data quickly while maintaining a clear focus on patient-centric objectives.
Challenges, responsibilities, and the road ahead
While the potential of AlphaFold and physics-based simulations is immense, Akinsanya emphasizes that challenges remain. Predictive models are powerful tools, but they are not infallible. Experimental validation remains essential, and the translation from computational predictions to clinical outcomes requires careful design, robust data, and rigorous regulatory considerations. Navigating these requirements while maintaining speed and efficiency is a central tension in modern drug development. The ethical and safety dimensions of AI-assisted discovery demand ongoing attention, including transparency in modeling approaches, reproducibility of results, and responsible data use.
Another frontier involves accessibility and equity in drug development. As these advanced computational capabilities become more widespread, it is important to ensure that small and mid-sized biotech groups, academic institutions, and diverse researchers can leverage them. Democratizing access to high-quality predictive tools can accelerate scientific progress across the globe and contribute to more equitable health outcomes. This broadened participation aligns with a shared commitment to reducing the burden of disease, particularly in settings where traditional resources for drug discovery may be limited.
Regulatory landscapes will also continue to evolve as computational methods become more integrated into the development pipeline. Establishing clear standards for how predictive models inform decision-making, how uncertainty is managed, and how computational evidence complements experimental data will be essential for efficient approvals and patient safety. The ongoing dialogue among scientists, clinicians, regulators, and industry leaders will shape how AlphaFold-informed strategies are implemented in practice, ensuring that innovation proceeds in a way that maintains public trust and maximizes therapeutic value.
In this context, the work at Schrödinger and similar research initiatives represents more than technological progress; it reflects a disciplined, patient-centered philosophy. The goal is to harness powerful computational tools to deliver medicines that improve outcomes for people facing cancer, heart disease, and related conditions. Achieving this vision requires not only sophisticated software and algorithms but also a deep commitment to scientific rigor, cross-disciplinary collaboration, and ethical responsibility. As researchers like Akinsanya continue to expand the capabilities of AlphaFold and companion simulations, the field moves closer to a future where the design of highly selective, safe, and effective therapies becomes a routine and reliable part of medical practice.
Implications for science, medicine, and society
The unfolding narrative around AlphaFold, physics-based modeling, and drug discovery signals a broader transformation in how science translates into medicine. The ability to understand receptor families with greater clarity and to predict how drugs interact within complex biological systems holds promise for more targeted therapies and improved patient outcomes. By focusing on selectivity and safety, researchers can reduce the risk of adverse effects while preserving or enhancing therapeutic efficacy. This shift has the potential to reshape clinical strategies, optimize treatment regimens, and expand the range of diseases that can be effectively addressed with rationally designed medicines.
From an educational standpoint, the convergence of AI, computational chemistry, and experimental biology offers rich opportunities for training and workforce development. The interdisciplinary nature of modern drug design requires new skill sets and collaborative practices, encouraging a new generation of scientists to think holistically about how molecules affect biological networks. Such cross-pollination can spur innovation across sectors, from academia to industry to healthcare, ultimately benefiting patients by expanding the pipeline of potential therapeutics.
In parallel, the broader public health implications are meaningful. As more drugs are designed with precision and safety in mind, the overall burden of cancer and heart disease could be alleviated, contributing to longer, healthier lives for patients. While no single technology guarantees rapid cures, the strategic use of AlphaFold in combination with physics-based simulations represents a substantial step toward more effective, patient-centered drug development. The ongoing collaboration among researchers, clinicians, and industry leaders will determine how quickly these advances translate into real-world benefits and how they are integrated into clinical practice to maximize impact.
Conclusion
In the evolving landscape of drug discovery, AlphaFold stands as a pivotal catalyst, enabling researchers to design more effective therapies with a level of precision that was once unattainable. Karen Akinsanya’s experience at Schrödinger highlights how a leadership perspective that bridges academia, industry, and clinical practice can accelerate the journey from bench to bedside. By leveraging AlphaFold’s structural predictions in tandem with physics-based simulations, researchers can explore how entire receptor families behave, identify subtle differences that drive selectivity, and design compounds with improved safety and efficacy profiles. This integrated approach promises to transform how medicines are developed for cancer, heart disease, and beyond, bringing us closer to the goal of delivering meaningful patient benefit more quickly and reliably. The ongoing commitment to rigorous validation, ethical responsibility, and collaborative innovation will shape the next era of therapeutic discovery, guided by the belief that nature’s complexity can be understood, harnessed, and translated into better health outcomes for patients around the world.