Atomic Precision in Personalized Oncology: AI-Designed Nanomedicines Enabling N-of-1 Cancer Therapy
More
Advanced Search
  • Open Access
  • Editors Choice
Perspective

Atomic Precision in Personalized Oncology: AI-Designed Nanomedicines Enabling N-of-1 Cancer Therapy
Click to copy article linkArticle link copied!

  • Miles Pourbaghi
    Miles Pourbaghi
    Department of Radiology, Weill Cornell Medicine, New York, New York 10021, United States
    More by Miles Pourbaghi
  • Michelle S. Bradbury*
    Michelle S. Bradbury
    Department of Radiology, Weill Cornell Medicine, New York, New York 10021, United States
    Sandra and Edward Meyer Cancer Center, Weill Cornell Medicine, New York, New York 10021, United States
    Department of Radiation Oncology, Weill Cornell Medicine, New York, New York 10065, United States
    *Email: [email protected]
    More by Michelle S. Bradbury
    Orcidhttps://orcid.org/0000-0003-3147-4391
Open PDF

ACS Nano Medicine

Cite this: ACS Nano Med. 2026, XXXX, XXX, XXX-XXX
Click to copy citationCitation copied!
https://doi.org/10.1021/acsnanomed.5c00178
Published April 6, 2026

© 2026 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Abstract

Click to copy section linkSection link copied!
High Resolution Image
Download MS PowerPoint Slide

Rapid advances in AI-driven molecular design, intelligent nanomaterial engineering, and clinical AI systems are reshaping personalized cancer therapy by enabling coordinated innovation across biological scales, from genome regulation to functional protein interactions. At the molecular level, generative platforms such as AlphaProteo and RFdiffusion now support the rapid design of de novo protein binders with high predicted structural accuracy and low-nanomolar affinities. Acting upstream, AlphaGenome interprets genome-wide regulatory variation to predict how single-base changes in noncoding DNA alter gene control mechanisms, facilitating the prioritization of patient-specific pathways and therapeutic targets. Together, these capabilities are transforming computational design from a speculative approach to a reliable molecular engineering workflow. In parallel, clinically validated nanomedicine platforms provide modular architectures that enhance pharmacokinetics, improve therapeutic indices, and modulate tumor–immune interactions, addressing persistent barriers in solid tumor treatment. Integration with AI-enabled experimental platforms, patient-derived organoid systems, spatial and single-cell profiling, quantitative systems pharmacology models, and microfluidic GMP-compatible manufacturing suggests a feasible path toward accelerated, patient-matched nanomedicine development. While challenges remain─including immunogenicity and manufacturing constraints─the convergence of computation, materials engineering, and regulatory science supports a realistic roadmap toward N-of-1 oncology.

This publication is licensed under

CC-BY-NC-ND 4.0 .
  • cc licence
  • by licence
  • nc licence
  • nd licence
© 2026 The Authors. Published by American Chemical Society

Subjects

what are subjects

Keywords

what are keywords
The convergence of AI-driven protein design and engineered nanomedicines marks an inflection point for personalized cancer therapy, where three previously distinct advances─computational protein design, intelligent nanomaterials engineering, and genomic regulation─now intersect to address long-standing barriers in refractory disease management. In this context, AI-enabled in silico protein design has dramatically shortened protein binder discovery timelines, enabling the generation of functional binders with high predicted structural accuracy on the scale of weeks rather than years. (1,2) Typical workflows begin with backbone generation using generative AI design methods, such as RFdiffusion, (1) followed by sequence and side-chain assignment with ProteinMPNN, (3) and structural validation using AlphaFold 3. (4) Together, these approaches enable a transition to the next frontier: accurate prediction and engineering of full protein–ligand interactions and multicomponent molecular assemblies (Figure 1).

Figure 1

Figure 1. N-of-1 convergence workflow: From biopsy to bedside. Schematic overview of a rapid-response-engineered therapeutic pipeline enabled by the convergence of AI-driven genomic interpretation and protein design. (A) Patient profiling: A tumor biopsy undergoes whole-genome sequencing and spatial transcriptomic analysis, with AlphaGenome applied to identify patient-specific regulatory variants and splice-derived antigenic peptides through systematic computational variant scanning. (B) AI-driven design: Generative platforms, including AlphaProteo and RFdiffusion, design de novo protein binders targeting the prioritized antigens, while AlphaFold 3 is used to assess predicted structural integrity and surface compatibility. (C) Rapid manufacturing: Automated microfluidic systems conjugate designed binders to an established, clinical-grade nanoparticle backbone (e.g., ultrasmall silica particles or lipid nanoparticles). (D) Ex vivo validation: The resulting patient-matched nanomedicine is evaluated in autologous, immune-competent tumor organoid systems to assess cellular uptake and potential immunogenicity. (E) Clinical administration: The validated N-of-1 therapeutic is administered, with quantitative systems pharmacology (QSP)-based digital twin models guiding dose selection.

High Resolution Image
Download MS PowerPoint Slide
In parallel, advances in precision-engineered nanomedicines are translating these computational gains into deliverable, clinically actionable agents. Nanomedicines, broadly defined as therapeutic formulations exploiting nanoscale size effects (typically 1–100 nm) to modulate pharmacokinetics, biodistribution, cellular uptake, and the tumor microenvironment (TME), offer distinct advantages over free drug administration. One notable example is a renally excreted and clinically validated ultrasmall silica nanoparticle, Cornell prime dots (C′ dots), (5,6) that modulates immune suppressive TMEs (7) and when conjugated with single-chain antibody fragments (scFvs (8)) yield favorable pharmacokinetic profiles and tumor-specific uptake. These findings address longstanding barriers in nanomedicine delivery, with preclinical studies reporting up to ∼17% tumor uptake without significant off-target accumulation. (8) Alongside these platforms, emerging technologies, including AI-guided closed-loop laboratories, (9) TME-responsive switches, (10) immune-competent organoids, (11) quantitative systems pharmacology (QSP) modeling, (12) AI-designed protein coronas, (13) and patient-specific digital twin screening, (14) are collectively positioning the field to approach true N-of-1 nanomedicine strategies.

The Inflection Point: From Anecdote to Systematic Engineering

Click to copy section linkSection link copied!
The period of 2024–2026 marked a definitive transition in molecular medicine, as computational approaches evolved from largely theoretical promise to operational reliability. While the 2024 Nobel Prize in Chemistry acknowledged foundational breakthroughs in protein structure determination, (15) the field’s current trajectory is increasingly defined by the accumulation of systematic validation data across three complementary pillars: genomic regulation, protein engineering, and materials science.
First, longstanding challenges in interpreting the regulatory code of the noncoding genome─the sequence-encoded rules that control gene expression, RNA processing, and chromatin state without altering protein-coding sequences─have been a major bottleneck in identifying N-of-1 targets. Published in January 2026, AlphaGenome represents a major advance in predicting the functional consequences of noncoding and splicing variants. By analyzing large genomic contexts spanning up to 1 million base pairs (Mb) with single-base resolution, the model matches or exceeds state-of-the-art performance in 25 of 26 standardized benchmarks. (16) Importantly, for personalized oncology, AlphaGenome applies systematic computational variant scanning, for instance, in silico saturation mutagenesis (ISM), to resolve the functional impact of rare pathogenic variants, including those affecting oncogenes such as TAL1. (16) This capability not only enables the prediction of whether a tumor harbors a given mutation but also how that mutation rewires the local regulatory landscape, thereby revealing patient-specific N-of-1 targets that are inaccessible to population-averaged approaches.
Second, independent experimental validation now supports the practical utility of the generative protein design. Moving beyond early proof-of-concept benchmarks, pipelines such as AlphaProteo have fundamentally altered binder development. AlphaProteo has demonstrated the ability to generate binders with pico- to nanomolar affinities across diverse targets, including proteins with limited prior structural characterization, achieving experimental success rates significantly higher than those of traditional high-throughput screening approaches. (17) This distinction is critical for clinical translation: generative models are now reliably delivering functional binding strength (Kd < 10 nM), transforming protein design from a largely stochastic process into a reproducible engineering workflow.
Third, the “intelligent nanomaterials” paradigm is similarly shifting from empirical screening toward generative design. In 2025, the introduction of MatterGen introduced a framework for creating novel, stable inorganic structures with predefined physical properties, such as magnetic or electronic constraints, directly from design prompts. (18) This advance moves the field closer to rationally engineering nanoparticle cores with the precisely tuned characteristics required for tumor penetration and biological performance rather than relying on incremental modification of existing material libraries.
In parallel, the pharmaceutical impact of AI is being quantified through rigorous pipeline analysis. A 2024 review revealed that AI-designed molecules have achieved Phase I clinical trial success rates of approximately 80–90%, compared with historical averages of 40–65%. (19) By contrast, Phase II efficacy success rates remain closer to historical norms (∼40%), underscoring that while AI has substantially improved early stage drug-likeness and safety prediction, forecasting complex biological efficacy remains a major hurdle that N-of-1 strategies must address through better patient stratification–precision where regulatory models like AlphaGenome will prove indispensable.

Why Nanoscale Architectures Are Uniquely Suited for N-of-1 Oncology

Click to copy section linkSection link copied!
Nanoscale architectures possess functional properties that cannot be replicated by small molecules or antibodies alone and are therefore uniquely suited for N-of-1 cancer therapy. Their tunable size, surface chemistry, and multivalent display enable simultaneous engagement of multiple patient-specific targets, including rare or subclonal antigens that may be inaccessible to single-epitope therapeutics. At sub-20 nm scales, nanoparticles exhibit substantially deeper tumor penetration compared to larger carriers and can achieve sufficient intratumoral retention (5,20) when surface chemistry and targeting are optimized, enabling them to navigate heterogeneous TMEs and access spatial niches that shape localized resistance patterns. Importantly, such platforms can serve as modular scaffolds supporting efficient conjugation of AI-designed binders, (21) stimulus-responsive release mechanisms, and logic-gated targeting switches (22) that enable therapeutic activation only when defined combinations of biological cues are present, all while preserving the core particle architecture and pharmacokinetic profile. Given their modularity, the same clinical-grade platform can be rapidly reconfigured with various patient-specific AI-designed targeting proteins, allowing the therapeutic payload to be tailored to individual molecular profiles. Collectively, these attributes position nanoscale architectures as a uniquely flexible and scalable foundation for individualized N-of-1 oncologic interventions.

Reimagining the Possible: Four Concepts across the Readiness Spectrum

Click to copy section linkSection link copied!
To rigorously chart the trajectory of personalized oncology, it is necessary to distinguish between technologies that are currently operational for early translation and those that represent emerging experimental concepts requiring further in vivo validation.

1. Digital Twin Discovery: From Patient Biopsy to Nanoparticle (Status: Operational in Early Translation)

The “Digital Twin” paradigm has been historically constrained by our limited ability to interpret the functional impact of patient-specific mutations. The integration of AlphaGenome (16) substantially alleviates this bottleneck. We propose a four week rapid-response loop: (i) Week 1 (Profile and Decode): Tumor biopsies undergo sequencing and spatial transcriptomic analysis, (23) subsequently interpreted by AlphaGenome. The model applies ISM to identify rare, patient-specific splice variants or regulatory mutations that generate novel surface antigenic peptides (neo-epitopes). Unlike previous models restricted to short sequences, AlphaGenome’s 1 Mb context window allows for the detection of distal regulatory effects that drive tumor heterogeneity. (ii) Week 2 (Design): Generative protein design platforms (e.g., AlphaProteo, RFdiffusion) create high-affinity binders (Kd < 10 nM) (1,24) targeting these validated neo-epitopes. (iii) Week 3 (Manufacture): Microfluidic conjugation to a clinical-grade backbone, followed by ex vivo validation. (25) This workflow transforms the “Digital Twin” from a retrospective model into a prospective molecular engineering engine, feasible within current clinical timelines.

2. Chameleon Coronas: Programming the “Bug” into a Feature (Status: Emerging Experimental Concept)

For more than two decades, the protein corona─the spontaneous adsorption of plasma proteins onto nanoparticle surfaces─has been viewed primarily as a barrier to targeting. Although the conceptual and potential biological feasibility of exploiting the protein corona in a personalized (N-of-1) context has been previously suggested, the use of AI was not reported. (26) We propose an alternative experimental paradigm in which AI-based predictive models are used to guide the modulation of the nanoparticle protein corona toward patient-specific targeting. While machine learning models can now predict corona composition with high fidelity in defined media, (13) the ability to rationally design surface motifs that selectively recruit functionality-enhancing host proteins in vivo remains an experimental frontier. The vision entails replacing inert stealth coatings with AI-designed interfaces that preferentially recruit dysregulated host proteins into a “self” biosignature that enhances tumor uptake.

3. Multispecific Protein Logic Gates: Boolean Therapeutics (Status: Emerging Experimental Concept)

Tumor heterogeneity and pathway redundancy frequently undermine single-agent therapies, as target expression, signaling dependencies, and microenvironmental cues vary spatially and temporally within and across lesions. We propose the development of “Boolean therapeutics” to address these limitations─nano-particle systems capable of executing logical operations (e.g., AND, OR, NOT) at the molecular scale to condition therapeutic activation on defined combinations of disease-associated signals (Figure 2). Recent advances have demonstrated computationally designed proteins that undergo conformational switching or functional activation upon ligand binding, enabling programmable responses to specific molecular inputs. (27) In parallel, stimulus-responsive nanomaterials have shown that drug release and imaging signals can be gated by environmental cues (pH, redox state, or enzymatic activity), establishing the feasibility of multi-input control in nanoscale systems. However, integrating these protein-based switches onto nanoparticle surfaces to perform coordinated, multi-input logic within the TME (e.g., payload release only if Target A AND Target B are present and local physicochemical conditions are permissive) is an emerging concept. Achieving reliable Boolean behavior in vivo will require careful control of ligand density, binding kinetics, and signal integration across heterogeneous cellular and stromal compartments as well as rigorous validation to ensure synchronized performance in heterogeneous biological settings.

Figure 2

Figure 2. Boolean therapeutics: Logic-gated nanomedicine design. Conceptual illustration of a multispecific “AND”-gate nanomedicine designed to address tumor heterogeneity while minimizing off-target toxicity. The nanoparticle is functionalized with two distinct AI-designed sensor proteins (Binder A and Binder B) and incorporates a pH-sensitive release mechanism. (Left) Healthy tissue: In the presence of only Target A (e.g., EGFR) and physiological pH, the therapeutic payload remains sequestered, preventing nonspecific release. (Right) Tumor microenvironment: Drug release is triggered only when the Boolean condition (Target A AND Target B) is satisfied in conjunction with low pH. This stringent logic requirement restricts payload delivery to tumor cells coexpressing defined antigens within the acidic tumor niche, thereby increasing the therapeutic index.

High Resolution Image
Download MS PowerPoint Slide

4. Self-Assembling Fractal Therapeutics: Maximizing Tumor Penetration (Status: Emerging Experimental Concept)

Solid tumors present substantial physical barriers to drug delivery including elevated interstitial pressure and a dense extracellular matrix, which collectively limit therapeutic penetration. To address this challenge, we explored the conceptual application of stimulus-responsive fractal assemblies. While the computational design of self-assembling protein fractals has been demonstrated in vitro, (28) extending this strategy to in vivo tumor delivery remains highly experimental. The proposed approach entails administering small, clinically clearable subunits that undergo environment-triggered self-assembly into larger, high-avidity fractal structures upon exposure to tumor-specific cues, conceptually analogous to size-transformable nanotherapeutics developed to reconcile deep tumor penetration with intratumoral retention. (29) In principle, this dynamic transformation could reconcile the long-standing “size paradox” of nanomedicine: Deep tumor penetration by small, diffusible subunits is followed by TME-triggered self-assembly into larger structures that enhance intratumoral retention. However, rigorous in vivo validation will be required to establish the feasibility and control.

The Path Forward: Timelines and Implementation Realities

Click to copy section linkSection link copied!
The transition from broad-spectrum chemotherapy to AI-designed, patient-specific nanomedicines is unlikely to occur as a single inflection point. Instead, it will unfold through a phased evolution driven by the convergence of computational design, materials engineering, and clinical translation (Figure 3). In the near term, progress is expected to shift from a computational proof of concept toward clinical safety and feasibility validation. AI-designed small molecules have already demonstrated Phase I success rates approaching 80–90%, (19) suggesting a comparable early trajectory for the first generation of AI-designed protein-nanoparticle conjugates. These initial efforts will likely rely on established, regulatoryly cleared particle backbones, such as silica or lipid-based carriers, functionalized with de novo binders targeting well-characterized antigens. At this stage, AI-guided closed-loop laboratories are anticipated to substantially shorten iterative optimization cycles, while immune-competent organoid systems provide a practical ex vivo platform for assessing binder specificity and functional activity.

Figure 3

Figure 3. Roadmap to N-of-1 oncology. Projected timeline distinguishing currently operational technologies from emerging experimental concepts. The roadmap illustrates the evolution of AI-designed nanomedicines across three translational phases: (1) Validation and safety (approximately 1–3 Years): Phase I evaluation of single-target nanoparticle conjugates and validation of immunogenicity prediction tools, with AlphaGenome operational for patient-specific target discovery and AlphaProteo supporting rapid binder design. (2) Integration and efficacy (approximately 3–5 Years): Deployment of “Digital Twin” QSP models for dose optimization, initial regulatory approvals of first-in-class AI-designed agents, and adoption of modular, microfluidic manufacturing platforms. (3) Complexity and precision (approximately 5–10 years): Routine clinical use of logic-gated nanotherapeutics and real-time, adaptive formulation cycles. Dashed lines denote technologies that currently require additional in vivo validation (e.g., fractal assemblies and higher-order logic-gated systems).

High Resolution Image
Download MS PowerPoint Slide
As these candidates advance, a medium-term phase is likely to emerge in which “digital twin” frameworks evolve from exploratory research tools into clinically relevant decision-support systems. Integration of QSP models─mechanistic frameworks that link drug exposure to biological response─with patient-specific multiomics data will become increasingly important for rational dose selection and treatment planning. (30) This period may coincide with the first regulatory approvals of AI-designed nanomedicines, supported by evolving guidance from agencies like the U.S. Food and Drug Administration (FDA) and European Medicines Agency (EMA) that increasingly recognize algorithmic and data-driven design paradigms. (31,32) In parallel, manufacturing strategies will need to transition toward continuous, modular microfluidic platforms capable of producing small-batch, on-demand formulations with appropriate quality control. (33)
Over longer horizons, the field may approach the routine deployment of fully integrated N-of-1 systems for refractory malignancies. In this more mature phase, tumor heterogeneity would be addressed not through serial trial and error, but through combinatorial, logic-gated nanomaterials designed in silico to match the specific subclonal architecture of an individual patient’s tumor. Real-time feedback loops could further refine these digital models during treatment, enabling the rapid synthesis of adaptive, “course-correcting” nanomedicines as the biological properties of the disease evolve.

Acknowledging the Obstacles

Click to copy section linkSection link copied!
Transitioning from population-averaged therapeutics to N-of-1 engineering requires a candid assessment of the scientific and regulatory barriers that remain. These challenges are not merely logistical but fundamental to the safety, robustness, and clinical feasibility of individualized therapies. One of the most immediate risks associated with AI-designed binders is immunogenicity. Although computational tools excel at stabilizing protein structures and optimizing binding interfaces, accurately predicting interactions with the adaptive immune system remains difficult. Systematic reviews indicate that antidrug antibody (ADA) incidence in engineered non-native protein scaffolds can range from near negligible to greater than 50%, depending on epitope content and structural context. (34) Unlike established monoclonal antibodies, which benefit from decades of clinical experience and well-characterized immunological behavior, de novo proteins introduce novel surface topologies with uncertain immune profiles. For this reason, integrating in silico MHC-II binding predictions together with rapid ex vivo T-cell assays into the design workflow should be regarded as a critical safety requirement rather than an optional refinement.
A second, less appreciated challenge arises from patient-specific variability in the nanoparticle protein corona. While predictive models of corona composition have achieved high accuracy under controlled experimental conditions, (13) they are typically trained on pooled human plasma data sets. In clinical settings, however, the plasma proteome is highly dynamic and individualized. A nanoparticle that preferentially recruits transport proteins in one patient may instead bind opsonins in another, particularly in the context of acute-phase inflammation, thereby profoundly altering the biodistribution, clearance, and efficacy. This interpatient variability limits the reliability of universal corona predictions and highlights the need to calibrate digital twin simulations using each patient’s specific proteomic and inflammatory profile.
Perhaps the most formidable obstacles are regulatory and manufacturing obstacles in nature. Current Good Manufacturing Practice (GMP) frameworks are predicated on the production of fixed and well-characterized products. N-of-1 therapies invert this paradigm; the therapeutic construct itself is inherently variable, changing from patient to patient, which renders traditional batch-based validation impractical. Enabling clinical translation will therefore require a shift toward process-based validation strategies in which the design, manufacturing, and quality-control pipelines are validated rather than individual products. In parallel, the manufacture of patient-specific, clinical-grade nanoparticle formulations on accelerated timelines will require new quality-control approaches, including rapid sterility and release assays that circumvent 14 day incubation requirements.

Conclusion: A Call to Convergence

Click to copy section linkSection link copied!
This moment calls for thinking beyond traditional disciplinary boundaries. Progress toward N-of-1 oncology will depend on the integration of computational protein design, materials engineering, systems modeling, clinical translation, and regulatory science into a cohesive, interoperable framework rather than a collection of parallel efforts. Many of the core capabilities required to design, validate, and manufacture patient-matched nanomedicines, including AlphaGenome for patient-specific target identification, AlphaProteo for rapid binder engineering, and MatterGen for materials discovery, are now largely operational. Consequently, the central challenge has shifted. The defining question is no longer whether these technologies can be developed in isolation but whether clinical workflows, manufacturing paradigms, and regulatory infrastructures can evolve in concert to support their responsible, scalable deployment.

Author Information

Click to copy section linkSection link copied!
  • Corresponding Author
    • Michelle S. Bradbury - Department of Radiology, Weill Cornell Medicine, New York, New York 10021, United States;  Sandra and Edward Meyer Cancer Center, Weill Cornell Medicine, New York, New York 10021, United States;  Department of Radiation Oncology, Weill Cornell Medicine, New York, New York 10065, United States;  Orcidhttps://orcid.org/0000-0003-3147-4391;  Email: [email protected]
  • Author
    • Miles Pourbaghi - Department of Radiology, Weill Cornell Medicine, New York, New York 10021, United States
  • Notes
    The authors declare no competing financial interest.

References

Click to copy section linkSection link copied!

This article references 34 other publications.

  1. 1
    Watson, J. L.; Juergens, D.; Bennett, N. R.; Trippe, B. L.; Yim, J.; Eisenach, H. E.; Ahern, W.; Borst, A. J.; Ragotte, R. J.; Milles, L. F. De novo design of protein structure and function with RFdiffusion. Nature 2023, 620 (7976), 1089– 1100,  DOI: 10.1038/s41586-023-06415-8
    Google Scholar
    There is no corresponding record for this reference.
  2. 2
    Fox, D. R.; Taveneau, C.; Clement, J.; Grinter, R.; Knott, G. J. Code to complex: AI-driven de novo binder design. Structure 2025, 33 (10), 1631– 1642,  DOI: 10.1016/j.str.2025.08.007
    Google Scholar
    There is no corresponding record for this reference.
  3. 3
    Dauparas, J.; Anishchenko, I.; Bennett, N.; Bai, H.; Ragotte, R. J.; Milles, L. F.; Wicky, B. I. M.; Courbet, A.; de Haas, R. J.; Bethel, N. Robust deep learning-based protein sequence design using ProteinMPNN. Science 2022, 378 (6615), 49– 56,  DOI: 10.1126/science.add2187
    Google Scholar
    There is no corresponding record for this reference.
  4. 4
    Abramson, J.; Adler, J.; Dunger, J.; Evans, R.; Green, T.; Pritzel, A.; Ronneberger, O.; Willmore, L.; Ballard, A. J.; Bambrick, J. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 2024, 630 (8016), 493– 500,  DOI: 10.1038/s41586-024-07487-w
    Google Scholar
    There is no corresponding record for this reference.
  5. 5
    Ma, K.; Mendoza, C.; Hanson, M.; Werner-Zwanziger, U.; Zwanziger, J.; Wiesner, U. Control of Ultrasmall Sub-10 nm Ligand-Functionalized Fluorescent Core-Shell Silica Nanoparticle Growth in Water. Chem. Mater. 2015, 27 (11), 4119– 4133,  DOI: 10.1021/acs.chemmater.5b01222
    Google Scholar
    There is no corresponding record for this reference.
  6. 6
    Phillips, E.; Penate-Medina, O.; Zanzonico, P. B.; Carvajal, R. D.; Mohan, P.; Ye, Y.; Humm, J.; Gönen, M.; Kalaigian, H.; Schöder, H. Clinical translation of an ultrasmall inorganic optical-PET imaging nanoparticle probe. Sci. Transl Med. 2014, 6 (260), 260ra149,  DOI: 10.1126/scitranslmed.3009524
    Google Scholar
    There is no corresponding record for this reference.
  7. 7
    De Leon, G.; Zhang, L.; Siddiqui, N. A.; Naguib, N.; Chen, F.; Padmanabhan, R.; Zhang, T.; Monette, S.; Socciarelli, F.; Lee, R. An ultrasmall core-shell silica nanoparticle improves antitumour immunity and survival by remodelling suppressive melanoma microenvironments. Nat. Nanotechnol. 2026, 21 (2), 311– 322,  DOI: 10.1038/s41565-025-02083-z
    Google Scholar
    There is no corresponding record for this reference.
  8. 8
    Chen, F.; Ma, K.; Madajewski, B.; Zhuang, L.; Zhang, L.; Rickert, K.; Marelli, M.; Yoo, B.; Turker, M. Z.; Overholtzer, M. Ultrasmall targeted nanoparticles with engineered antibody fragments for imaging detection of HER2-overexpressing breast cancer. Nat. Commun. 2018, 9 (1), 4141,  DOI: 10.1038/s41467-018-06271-5
    Google Scholar
    There is no corresponding record for this reference.
  9. 9
    Szymanski, N. J.; Rendy, B.; Fei, Y.; Kumar, R. E.; He, T.; Milsted, D.; McDermott, M. J.; Gallant, M.; Cubuk, E. D.; Merchant, A. An autonomous laboratory for the accelerated synthesis of novel materials. Nature 2023, 624 (7990), 86– 91,  DOI: 10.1038/s41586-023-06734-w
    Google Scholar
    There is no corresponding record for this reference.
  10. 10
    Velapure, P.; Kansal, D.; Bobade, C. Tumor microenvironment-responsive nanoformulations for breast cancer. Discov Nano 2024, 19 (1), 212,  DOI: 10.1186/s11671-024-04122-5
    Google Scholar
    There is no corresponding record for this reference.
  11. 11
    Wang, J.; Tao, X.; Zhu, J.; Dai, Z.; Du, Y.; Xie, Y.; Chu, X.; Fu, G.; Lei, Z. Tumor organoid-immune co-culture models: exploring a new perspective of tumor immunity. Cell Death Discov 2025, 11 (1), 195,  DOI: 10.1038/s41420-025-02407-x
    Google Scholar
    There is no corresponding record for this reference.
  12. 12
    Helmlinger, G.; Sokolov, V.; Peskov, K.; Hallow, K. M.; Kosinsky, Y.; Voronova, V.; Chu, L.; Yakovleva, T.; Azarov, I.; Kaschek, D. Quantitative Systems Pharmacology: An Exemplar Model-Building Workflow With Applications in Cardiovascular, Metabolic, and Oncology Drug Development. CPT Pharmacometrics Syst. Pharmacol 2019, 8 (6), 380– 395,  DOI: 10.1002/psp4.12426
    Google Scholar
    There is no corresponding record for this reference.
  13. 13
    Canchola, A.; Li, K.; Chen, K.; Borboa-Pimentel, A.; Chou, C.; Dela Rama, R.; Chen, C. Y.; Chen, X.; Strobel, M.; Riviere, J. E. Meta-Analysis and Machine Learning Prediction of Protein Corona Composition across Nanoparticle Systems in Biological Media. ACS Nano 2025, 19, 37633,  DOI: 10.1021/acsnano.5c08608
    Google Scholar
    There is no corresponding record for this reference.
  14. 14
    Wang, H.; Arulraj, T.; Ippolito, A.; Popel, A. S. From virtual patients to digital twins in immuno-oncology: lessons learned from mechanistic quantitative systems pharmacology modeling. NPJ. Digit Med. 2024, 7 (1), 189,  DOI: 10.1038/s41746-024-01188-4
    Google Scholar
    There is no corresponding record for this reference.
  15. 15
    Nobel Prize in Chemistry 2024 , 2024. https://www.nobelprize.org/uploads/2024/10/press-chemistryprize2024-3.pdf.
    Google Scholar
    There is no corresponding record for this reference.
  16. 16
    Avsec, Ž.; Latysheva, N.; Cheng, J.; Novati, G.; Taylor, K. R.; Ward, T.; Bycroft, C.; Nicolaisen, L.; Arvaniti, E.; Pan, J. Advancing regulatory variant effect prediction with AlphaGenome. Nature 2026, 649 (8099), 1206– 1218,  DOI: 10.1038/s41586-025-10014-0
    Google Scholar
    There is no corresponding record for this reference.
  17. 17
    Zambaldi, V.; La, D.; Chu, A. E.; Patani, H.; Danson, A. E.; Kwan, T. O.; Frerix, T.; Schneider, R. G.; Saxton, D.; Thillaisundaram, A. De novo design of high-affinity protein binders with AlphaProteo. arXiv Preprint , arXiv:2409.08022, 2024.
    Google Scholar
    There is no corresponding record for this reference.
  18. 18
    Zeni, C.; Pinsler, R.; Zügner, D.; Fowler, A.; Horton, M.; Fu, X.; Wang, Z.; Shysheya, A.; Crabbé, J.; Ueda, S. A generative model for inorganic materials design. Nature 2025, 639 (8055), 624– 632,  DOI: 10.1038/s41586-025-08628-5
    Google Scholar
    There is no corresponding record for this reference.
  19. 19
    Kp Jayatunga, M.; Ayers, M.; Bruens, L.; Jayanth, D.; Meier, C. How successful are AI-discovered drugs in clinical trials? A first analysis and emerging lessons. Drug Discov Today 2024, 29 (6), 104009,  DOI: 10.1016/j.drudis.2024.104009
    Google Scholar
    There is no corresponding record for this reference.
  20. 20
    Lee, K. Y.; Lee, G. Y.; Lane, L. A.; Li, B.; Wang, J.; Lu, Q.; Wang, Y.; Nie, S. Functionalized, Long-Circulating, and Ultrasmall Gold Nanocarriers for Overcoming the Barriers of Low Nanoparticle Delivery Efficiency and Poor Tumor Penetration. Bioconjug Chem. 2017, 28 (1), 244– 252,  DOI: 10.1021/acs.bioconjchem.6b00224
    Google Scholar
    There is no corresponding record for this reference.
  21. 21
    Broske, B.; Binder, S. C.; McEnroe, B. A.; Kempchen, T. N.; Fandrey, C. I.; Messmer, J. M.; Tan, E.; Konopka, P.; Ferber, D.; Yong, M. C. R. Development of AI-designed protein binders for detection and targeting of cancer cell surface proteins. bioRxiv Preprint , 2025.  DOI: 10.1101/2025.05.11.652819 .
    Google Scholar
    There is no corresponding record for this reference.
  22. 22
    Chen, Z.; Kibler, R. D.; Hunt, A.; Busch, F.; Pearl, J.; Jia, M.; VanAernum, Z. L.; Wicky, B. I. M.; Dods, G.; Liao, H. De novo design of protein logic gates. Science 2020, 368 (6486), 78– 84,  DOI: 10.1126/science.aay2790
    Google Scholar
    There is no corresponding record for this reference.
  23. 23
    Johnson, A.; Shen, Y.; Zheng, X.; Su, F.; Zeng, J.; Uzunparmak, B.; Breuer, E.-K.; Holla, V.; Kahle, M.; Kim, S.-H. The actionable transcriptome: a framework for incorporating RNA sequencing into precision oncology. Nature Reviews Clinical Oncology 2026, 23, 213,  DOI: 10.1038/s41571-025-01110-1
    Google Scholar
    There is no corresponding record for this reference.
  24. 24
    Bennett, N. R.; Coventry, B.; Goreshnik, I.; Huang, B.; Allen, A.; Vafeados, D.; Peng, Y. P.; Dauparas, J.; Baek, M.; Stewart, L. Improving de novo protein binder design with deep learning. Nat. Commun. 2023, 14 (1), 2625,  DOI: 10.1038/s41467-023-38328-5
    Google Scholar
    There is no corresponding record for this reference.
  25. 25
    Valencia, P. M.; Pridgen, E. M.; Rhee, M.; Langer, R.; Farokhzad, O. C.; Karnik, R. Microfluidic platform for combinatorial synthesis and optimization of targeted nanoparticles for cancer therapy. ACS Nano 2013, 7 (12), 10671– 10680,  DOI: 10.1021/nn403370e
    Google Scholar
    There is no corresponding record for this reference.
  26. 26
    Corbo, C.; Molinaro, R.; Tabatabaei, M.; Farokhzad, O. C.; Mahmoudi, M. Personalized protein corona on nanoparticles and its clinical implications. Biomater Sci. 2017, 5 (3), 378– 387,  DOI: 10.1039/C6BM00921B
    Google Scholar
    There is no corresponding record for this reference.
  27. 27
    Praetorius, F.; Leung, P. J. Y.; Tessmer, M. H.; Broerman, A.; Demakis, C.; Dishman, A. F.; Pillai, A.; Idris, A.; Juergens, D.; Dauparas, J. Design of stimulus-responsive two-state hinge proteins. Science 2023, 381 (6659), 754– 760,  DOI: 10.1126/science.adg7731
    Google Scholar
    There is no corresponding record for this reference.
  28. 28
    Hsia, Y.; Bale, J. B.; Gonen, S.; Shi, D.; Sheffler, W.; Fong, K. K.; Nattermann, U.; Xu, C.; Huang, P. S.; Ravichandran, R. Design of a hyperstable 60-subunit protein dodecahedron. [corrected]. Nature 2016, 535 (7610), 136– 139,  DOI: 10.1038/nature18010
    Google Scholar
    There is no corresponding record for this reference.
  29. 29
    Ma, T.; Tran, T. B.; Lin, E.; Hunt, S.; Haveman, R.; Castro, K.; Lu, J. Size-transformable nanotherapeutics for cancer therapy. Acta Pharm. Sin B 2025, 15 (2), 834– 851,  DOI: 10.1016/j.apsb.2024.11.012
    Google Scholar
    There is no corresponding record for this reference.
  30. 30
    Joyce, P.; Allen, C. J.; Alonso, M. J.; Ashford, M.; Bradbury, M. S.; Germain, M.; Kavallaris, M.; Langer, R.; Lammers, T.; Peracchia, M. T. A translational framework to DELIVER nanomedicines to the clinic. Nat. Nanotechnol 2024, 19 (11), 1597– 1611,  DOI: 10.1038/s41565-024-01754-7
    Google Scholar
    There is no corresponding record for this reference.
  31. 31
    Considerations for the Use of Artificial Intelligence To Support Regulatory Decision-Making for Drug and Biological Products , 2025. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-use-artificial-intelligence-support-regulatory-decision-making-drug-and-biological.
    Google Scholar
    There is no corresponding record for this reference.
  32. 32
    Reflection paper on the use of Artificial Intelligence (AI) in the medicinal product lifecycle , 2024. https://www.ema.europa.eu/en/documents/scientific-guideline/reflection-paper-use-artificial-intelligence-ai-medicinal-product-lifecycle_en.pdf.
    Google Scholar
    There is no corresponding record for this reference.
  33. 33
    Sharma, V.; Mottafegh, A.; Joo, J. U.; Kang, J. H.; Wang, L.; Kim, D. P. Toward microfluidic continuous-flow and intelligent downstream processing of biopharmaceuticals. Lab Chip 2024, 24 (11), 2861– 2882,  DOI: 10.1039/D3LC01097J
    Google Scholar
    There is no corresponding record for this reference.
  34. 34
    Carter, P. J.; Quarmby, V. Immunogenicity risk assessment and mitigation for engineered antibody and protein therapeutics. Nat. Rev. Drug Discov 2024, 23 (12), 898– 913,  DOI: 10.1038/s41573-024-01051-x
    Google Scholar
    There is no corresponding record for this reference.

Cited By

Click to copy section linkSection link copied!

This article has not yet been cited by other publications.

Go to
Get e-Alerts
Get e-Alerts

ACS Nano Medicine

Cite this: ACS Nano Med. 2026, XXXX, XXX, XXX-XXX
Click to copy citationCitation copied!
https://doi.org/10.1021/acsnanomed.5c00178
Published April 6, 2026

© 2026 The Authors. Published by American Chemical Society. This publication is licensed under

CC-BY-NC-ND 4.0 .

Article Views

132

Altmetric

-

Citations

-
Learn about these metrics

Article Views are the COUNTER-compliant sum of full text article downloads since November 2008 (both PDF and HTML) across all institutions and individuals. These metrics are regularly updated to reflect usage leading up to the last few days.

Citations are the number of other articles citing this article, calculated by Crossref and updated daily. Find more information about Crossref citation counts.

The Altmetric Attention Score is a quantitative measure of the attention that a research article has received online. Clicking on the donut icon will load a page at altmetric.com with additional details about the score and the social media presence for the given article. Find more information on the Altmetric Attention Score and how the score is calculated.

  • Abstract

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 1

    Figure 1. N-of-1 convergence workflow: From biopsy to bedside. Schematic overview of a rapid-response-engineered therapeutic pipeline enabled by the convergence of AI-driven genomic interpretation and protein design. (A) Patient profiling: A tumor biopsy undergoes whole-genome sequencing and spatial transcriptomic analysis, with AlphaGenome applied to identify patient-specific regulatory variants and splice-derived antigenic peptides through systematic computational variant scanning. (B) AI-driven design: Generative platforms, including AlphaProteo and RFdiffusion, design de novo protein binders targeting the prioritized antigens, while AlphaFold 3 is used to assess predicted structural integrity and surface compatibility. (C) Rapid manufacturing: Automated microfluidic systems conjugate designed binders to an established, clinical-grade nanoparticle backbone (e.g., ultrasmall silica particles or lipid nanoparticles). (D) Ex vivo validation: The resulting patient-matched nanomedicine is evaluated in autologous, immune-competent tumor organoid systems to assess cellular uptake and potential immunogenicity. (E) Clinical administration: The validated N-of-1 therapeutic is administered, with quantitative systems pharmacology (QSP)-based digital twin models guiding dose selection.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 2

    Figure 2. Boolean therapeutics: Logic-gated nanomedicine design. Conceptual illustration of a multispecific “AND”-gate nanomedicine designed to address tumor heterogeneity while minimizing off-target toxicity. The nanoparticle is functionalized with two distinct AI-designed sensor proteins (Binder A and Binder B) and incorporates a pH-sensitive release mechanism. (Left) Healthy tissue: In the presence of only Target A (e.g., EGFR) and physiological pH, the therapeutic payload remains sequestered, preventing nonspecific release. (Right) Tumor microenvironment: Drug release is triggered only when the Boolean condition (Target A AND Target B) is satisfied in conjunction with low pH. This stringent logic requirement restricts payload delivery to tumor cells coexpressing defined antigens within the acidic tumor niche, thereby increasing the therapeutic index.

    High Resolution Image
    Download MS PowerPoint Slide

    Figure 3

    Figure 3. Roadmap to N-of-1 oncology. Projected timeline distinguishing currently operational technologies from emerging experimental concepts. The roadmap illustrates the evolution of AI-designed nanomedicines across three translational phases: (1) Validation and safety (approximately 1–3 Years): Phase I evaluation of single-target nanoparticle conjugates and validation of immunogenicity prediction tools, with AlphaGenome operational for patient-specific target discovery and AlphaProteo supporting rapid binder design. (2) Integration and efficacy (approximately 3–5 Years): Deployment of “Digital Twin” QSP models for dose optimization, initial regulatory approvals of first-in-class AI-designed agents, and adoption of modular, microfluidic manufacturing platforms. (3) Complexity and precision (approximately 5–10 years): Routine clinical use of logic-gated nanotherapeutics and real-time, adaptive formulation cycles. Dashed lines denote technologies that currently require additional in vivo validation (e.g., fractal assemblies and higher-order logic-gated systems).

    High Resolution Image
    Download MS PowerPoint Slide

  • References

    This article references 34 other publications.

    1. 1
      Watson, J. L.; Juergens, D.; Bennett, N. R.; Trippe, B. L.; Yim, J.; Eisenach, H. E.; Ahern, W.; Borst, A. J.; Ragotte, R. J.; Milles, L. F. De novo design of protein structure and function with RFdiffusion. Nature 2023, 620 (7976), 1089– 1100,  DOI: 10.1038/s41586-023-06415-8
      There is no corresponding record for this reference.
    2. 2
      Fox, D. R.; Taveneau, C.; Clement, J.; Grinter, R.; Knott, G. J. Code to complex: AI-driven de novo binder design. Structure 2025, 33 (10), 1631– 1642,  DOI: 10.1016/j.str.2025.08.007
      There is no corresponding record for this reference.
    3. 3
      Dauparas, J.; Anishchenko, I.; Bennett, N.; Bai, H.; Ragotte, R. J.; Milles, L. F.; Wicky, B. I. M.; Courbet, A.; de Haas, R. J.; Bethel, N. Robust deep learning-based protein sequence design using ProteinMPNN. Science 2022, 378 (6615), 49– 56,  DOI: 10.1126/science.add2187
      There is no corresponding record for this reference.
    4. 4
      Abramson, J.; Adler, J.; Dunger, J.; Evans, R.; Green, T.; Pritzel, A.; Ronneberger, O.; Willmore, L.; Ballard, A. J.; Bambrick, J. Accurate structure prediction of biomolecular interactions with AlphaFold 3. Nature 2024, 630 (8016), 493– 500,  DOI: 10.1038/s41586-024-07487-w
      There is no corresponding record for this reference.
    5. 5
      Ma, K.; Mendoza, C.; Hanson, M.; Werner-Zwanziger, U.; Zwanziger, J.; Wiesner, U. Control of Ultrasmall Sub-10 nm Ligand-Functionalized Fluorescent Core-Shell Silica Nanoparticle Growth in Water. Chem. Mater. 2015, 27 (11), 4119– 4133,  DOI: 10.1021/acs.chemmater.5b01222
      There is no corresponding record for this reference.
    6. 6
      Phillips, E.; Penate-Medina, O.; Zanzonico, P. B.; Carvajal, R. D.; Mohan, P.; Ye, Y.; Humm, J.; Gönen, M.; Kalaigian, H.; Schöder, H. Clinical translation of an ultrasmall inorganic optical-PET imaging nanoparticle probe. Sci. Transl Med. 2014, 6 (260), 260ra149,  DOI: 10.1126/scitranslmed.3009524
      There is no corresponding record for this reference.
    7. 7
      De Leon, G.; Zhang, L.; Siddiqui, N. A.; Naguib, N.; Chen, F.; Padmanabhan, R.; Zhang, T.; Monette, S.; Socciarelli, F.; Lee, R. An ultrasmall core-shell silica nanoparticle improves antitumour immunity and survival by remodelling suppressive melanoma microenvironments. Nat. Nanotechnol. 2026, 21 (2), 311– 322,  DOI: 10.1038/s41565-025-02083-z
      There is no corresponding record for this reference.
    8. 8
      Chen, F.; Ma, K.; Madajewski, B.; Zhuang, L.; Zhang, L.; Rickert, K.; Marelli, M.; Yoo, B.; Turker, M. Z.; Overholtzer, M. Ultrasmall targeted nanoparticles with engineered antibody fragments for imaging detection of HER2-overexpressing breast cancer. Nat. Commun. 2018, 9 (1), 4141,  DOI: 10.1038/s41467-018-06271-5
      There is no corresponding record for this reference.
    9. 9
      Szymanski, N. J.; Rendy, B.; Fei, Y.; Kumar, R. E.; He, T.; Milsted, D.; McDermott, M. J.; Gallant, M.; Cubuk, E. D.; Merchant, A. An autonomous laboratory for the accelerated synthesis of novel materials. Nature 2023, 624 (7990), 86– 91,  DOI: 10.1038/s41586-023-06734-w
      There is no corresponding record for this reference.
    10. 10
      Velapure, P.; Kansal, D.; Bobade, C. Tumor microenvironment-responsive nanoformulations for breast cancer. Discov Nano 2024, 19 (1), 212,  DOI: 10.1186/s11671-024-04122-5
      There is no corresponding record for this reference.
    11. 11
      Wang, J.; Tao, X.; Zhu, J.; Dai, Z.; Du, Y.; Xie, Y.; Chu, X.; Fu, G.; Lei, Z. Tumor organoid-immune co-culture models: exploring a new perspective of tumor immunity. Cell Death Discov 2025, 11 (1), 195,  DOI: 10.1038/s41420-025-02407-x
      There is no corresponding record for this reference.
    12. 12
      Helmlinger, G.; Sokolov, V.; Peskov, K.; Hallow, K. M.; Kosinsky, Y.; Voronova, V.; Chu, L.; Yakovleva, T.; Azarov, I.; Kaschek, D. Quantitative Systems Pharmacology: An Exemplar Model-Building Workflow With Applications in Cardiovascular, Metabolic, and Oncology Drug Development. CPT Pharmacometrics Syst. Pharmacol 2019, 8 (6), 380– 395,  DOI: 10.1002/psp4.12426
      There is no corresponding record for this reference.
    13. 13
      Canchola, A.; Li, K.; Chen, K.; Borboa-Pimentel, A.; Chou, C.; Dela Rama, R.; Chen, C. Y.; Chen, X.; Strobel, M.; Riviere, J. E. Meta-Analysis and Machine Learning Prediction of Protein Corona Composition across Nanoparticle Systems in Biological Media. ACS Nano 2025, 19, 37633,  DOI: 10.1021/acsnano.5c08608
      There is no corresponding record for this reference.
    14. 14
      Wang, H.; Arulraj, T.; Ippolito, A.; Popel, A. S. From virtual patients to digital twins in immuno-oncology: lessons learned from mechanistic quantitative systems pharmacology modeling. NPJ. Digit Med. 2024, 7 (1), 189,  DOI: 10.1038/s41746-024-01188-4
      There is no corresponding record for this reference.
    15. 15
      Nobel Prize in Chemistry 2024 , 2024. https://www.nobelprize.org/uploads/2024/10/press-chemistryprize2024-3.pdf.
      There is no corresponding record for this reference.
    16. 16
      Avsec, Ž.; Latysheva, N.; Cheng, J.; Novati, G.; Taylor, K. R.; Ward, T.; Bycroft, C.; Nicolaisen, L.; Arvaniti, E.; Pan, J. Advancing regulatory variant effect prediction with AlphaGenome. Nature 2026, 649 (8099), 1206– 1218,  DOI: 10.1038/s41586-025-10014-0
      There is no corresponding record for this reference.
    17. 17
      Zambaldi, V.; La, D.; Chu, A. E.; Patani, H.; Danson, A. E.; Kwan, T. O.; Frerix, T.; Schneider, R. G.; Saxton, D.; Thillaisundaram, A. De novo design of high-affinity protein binders with AlphaProteo. arXiv Preprint , arXiv:2409.08022, 2024.
      There is no corresponding record for this reference.
    18. 18
      Zeni, C.; Pinsler, R.; Zügner, D.; Fowler, A.; Horton, M.; Fu, X.; Wang, Z.; Shysheya, A.; Crabbé, J.; Ueda, S. A generative model for inorganic materials design. Nature 2025, 639 (8055), 624– 632,  DOI: 10.1038/s41586-025-08628-5
      There is no corresponding record for this reference.
    19. 19
      Kp Jayatunga, M.; Ayers, M.; Bruens, L.; Jayanth, D.; Meier, C. How successful are AI-discovered drugs in clinical trials? A first analysis and emerging lessons. Drug Discov Today 2024, 29 (6), 104009,  DOI: 10.1016/j.drudis.2024.104009
      There is no corresponding record for this reference.
    20. 20
      Lee, K. Y.; Lee, G. Y.; Lane, L. A.; Li, B.; Wang, J.; Lu, Q.; Wang, Y.; Nie, S. Functionalized, Long-Circulating, and Ultrasmall Gold Nanocarriers for Overcoming the Barriers of Low Nanoparticle Delivery Efficiency and Poor Tumor Penetration. Bioconjug Chem. 2017, 28 (1), 244– 252,  DOI: 10.1021/acs.bioconjchem.6b00224
      There is no corresponding record for this reference.
    21. 21
      Broske, B.; Binder, S. C.; McEnroe, B. A.; Kempchen, T. N.; Fandrey, C. I.; Messmer, J. M.; Tan, E.; Konopka, P.; Ferber, D.; Yong, M. C. R. Development of AI-designed protein binders for detection and targeting of cancer cell surface proteins. bioRxiv Preprint , 2025.  DOI: 10.1101/2025.05.11.652819 .
      There is no corresponding record for this reference.
    22. 22
      Chen, Z.; Kibler, R. D.; Hunt, A.; Busch, F.; Pearl, J.; Jia, M.; VanAernum, Z. L.; Wicky, B. I. M.; Dods, G.; Liao, H. De novo design of protein logic gates. Science 2020, 368 (6486), 78– 84,  DOI: 10.1126/science.aay2790
      There is no corresponding record for this reference.
    23. 23
      Johnson, A.; Shen, Y.; Zheng, X.; Su, F.; Zeng, J.; Uzunparmak, B.; Breuer, E.-K.; Holla, V.; Kahle, M.; Kim, S.-H. The actionable transcriptome: a framework for incorporating RNA sequencing into precision oncology. Nature Reviews Clinical Oncology 2026, 23, 213,  DOI: 10.1038/s41571-025-01110-1
      There is no corresponding record for this reference.
    24. 24
      Bennett, N. R.; Coventry, B.; Goreshnik, I.; Huang, B.; Allen, A.; Vafeados, D.; Peng, Y. P.; Dauparas, J.; Baek, M.; Stewart, L. Improving de novo protein binder design with deep learning. Nat. Commun. 2023, 14 (1), 2625,  DOI: 10.1038/s41467-023-38328-5
      There is no corresponding record for this reference.
    25. 25
      Valencia, P. M.; Pridgen, E. M.; Rhee, M.; Langer, R.; Farokhzad, O. C.; Karnik, R. Microfluidic platform for combinatorial synthesis and optimization of targeted nanoparticles for cancer therapy. ACS Nano 2013, 7 (12), 10671– 10680,  DOI: 10.1021/nn403370e
      There is no corresponding record for this reference.
    26. 26
      Corbo, C.; Molinaro, R.; Tabatabaei, M.; Farokhzad, O. C.; Mahmoudi, M. Personalized protein corona on nanoparticles and its clinical implications. Biomater Sci. 2017, 5 (3), 378– 387,  DOI: 10.1039/C6BM00921B
      There is no corresponding record for this reference.
    27. 27
      Praetorius, F.; Leung, P. J. Y.; Tessmer, M. H.; Broerman, A.; Demakis, C.; Dishman, A. F.; Pillai, A.; Idris, A.; Juergens, D.; Dauparas, J. Design of stimulus-responsive two-state hinge proteins. Science 2023, 381 (6659), 754– 760,  DOI: 10.1126/science.adg7731
      There is no corresponding record for this reference.
    28. 28
      Hsia, Y.; Bale, J. B.; Gonen, S.; Shi, D.; Sheffler, W.; Fong, K. K.; Nattermann, U.; Xu, C.; Huang, P. S.; Ravichandran, R. Design of a hyperstable 60-subunit protein dodecahedron. [corrected]. Nature 2016, 535 (7610), 136– 139,  DOI: 10.1038/nature18010
      There is no corresponding record for this reference.
    29. 29
      Ma, T.; Tran, T. B.; Lin, E.; Hunt, S.; Haveman, R.; Castro, K.; Lu, J. Size-transformable nanotherapeutics for cancer therapy. Acta Pharm. Sin B 2025, 15 (2), 834– 851,  DOI: 10.1016/j.apsb.2024.11.012
      There is no corresponding record for this reference.
    30. 30
      Joyce, P.; Allen, C. J.; Alonso, M. J.; Ashford, M.; Bradbury, M. S.; Germain, M.; Kavallaris, M.; Langer, R.; Lammers, T.; Peracchia, M. T. A translational framework to DELIVER nanomedicines to the clinic. Nat. Nanotechnol 2024, 19 (11), 1597– 1611,  DOI: 10.1038/s41565-024-01754-7
      There is no corresponding record for this reference.
    31. 31
      Considerations for the Use of Artificial Intelligence To Support Regulatory Decision-Making for Drug and Biological Products , 2025. https://www.fda.gov/regulatory-information/search-fda-guidance-documents/considerations-use-artificial-intelligence-support-regulatory-decision-making-drug-and-biological.
      There is no corresponding record for this reference.
    32. 32
      Reflection paper on the use of Artificial Intelligence (AI) in the medicinal product lifecycle , 2024. https://www.ema.europa.eu/en/documents/scientific-guideline/reflection-paper-use-artificial-intelligence-ai-medicinal-product-lifecycle_en.pdf.
      There is no corresponding record for this reference.
    33. 33
      Sharma, V.; Mottafegh, A.; Joo, J. U.; Kang, J. H.; Wang, L.; Kim, D. P. Toward microfluidic continuous-flow and intelligent downstream processing of biopharmaceuticals. Lab Chip 2024, 24 (11), 2861– 2882,  DOI: 10.1039/D3LC01097J
      There is no corresponding record for this reference.
    34. 34
      Carter, P. J.; Quarmby, V. Immunogenicity risk assessment and mitigation for engineered antibody and protein therapeutics. Nat. Rev. Drug Discov 2024, 23 (12), 898– 913,  DOI: 10.1038/s41573-024-01051-x
      There is no corresponding record for this reference.