Computationally guided high-throughput design of self-assembling drug nanoparticles

Advanced nanoparticle engineering for precision therapeutics of brain diseases

Despite the increasing global prevalence of neurological disorders, the development of nanoparticle (NP) technologies for brain-targeted therapies confronts considerable challenges. One of the key obstacles in treating brain diseases is the blood-brain barrier (BBB), which restricts the penetration of NP-based therapies into the brain. To address this issue, NPs can be installed with specific ligands or bioengineered to boost their precision and efficacy in targeting brain-diseased cells by navigating across the BBB, ultimately improving patient treatment outcomes. At the outset of this review, we highlighted the critical role of ligand-functionalized or bioengineered NPs in treating brain diseases from a clinical perspective. We then identified the key obstacles and challenges NPs encounter during brain delivery, including immune clearance, capture by the reticuloendothelial system (RES), the BBB, and the complex post-BBB microenvironment. Following this, we overviewed the recent progress in NPs engineering, focusing on ligand-functionalization or bionic designs to enable active BBB transcytosis and targeted delivery to brain-diseased cells. Lastly, we summarized the critical challenges hindering clinical translation, including scalability issues and off-target effects, while outlining future opportunities for designing cutting-edge brain delivery technologies.

Antioxidant and antimicrobial sodium alginate/sodium carboxymethyl cellulose films loaded with self-assembled hesperidin nanorods for fruits preservation

To develop green food packaging films, we prepared carrier-free hesperidin (HSD) nanoparticles by self-assembly technique and loaded them into Sodium Alginate (SA) and Sodium Carboxymethyl Cellulose (CMC) matrices to obtain the composite film. SEM and TEM imaging revealed that these nanoparticles exhibited a rod-like structure (hesperidin nanorods, HSD NRs). Compared to HSD, the water solubility and biological activity of HSD NRs were significantly higher. When HSD NRs were loaded into the SA/CMC film, the antibacterial ratios against S. aureus and E. coli were 91.20 % and 78.02 %, respectively. Moreover, the composite film showed good antioxidant activity against DPPH+, ABTS+, and Fe3+. The addition of HSD NRs significantly improved the film's barrier properties and mechanical strength. Therefore, the SA/CMC-HSD NRs film was more effective than the traditional polyethylene (PE) film in extending the shelf life. These results indicate that the SA/CMC-HSD NRs film holds great potential for fruit preservation.

Breaking the barriers in effective and safe toll-like receptor stimulation via nano-immunomodulators for potent cancer immunotherapy

Immunotherapy is an emerging strategy that awakens the intrinsic immune system for cancer treatment. Generally, successful immunotherapy of malignant tumours relies on the effective production of tumour-associated antigens and their lymph node delivery, antigen processing and presentation for T-cell activation, and the dismantling of the immunosuppressive tumour microenvironment. Toll-like receptor (TLR) agonists are potent stimulants in cancer immunotherapy, which can directly activate antigen-presenting cells (APCs) and further induce T cell activation for antitumour immune response and convert immunosuppressive tumour microenvironment to an immunogenic one for cooperative tumour ablation. However, TLR agonists for effective cancer immunotherapy have encountered essential challenges, such as insufficient immune activation and systemic side effects. In recent years, nano-immunomodulators with TLR agonists have been employed for tumour- and/or lymph node-targeted immune activation to improve the antitumour immune response and alleviate their systemic toxicities, providing a promising strategy for enhanced cancer immunotherapy. Herein, we introduce the recent progress in developing various TLR nano-immunomodulators for cancer immunotherapy via APCs activation and tumour microenvironment remodelling. Upon elucidating the rational design principles of nano-immunomodulators, we elucidate the advancement of TLR nanoagonists to break the barriers in effective and safe Toll-like receptor stimulation for potent cancer immunotherapy.

Mastering the complexities of cancer nanomedicine with text mining, AI and automation

In this contribution to the Orations - New Horizons of the Journal of Controlled Release, I present a personal perspective on the complexities of cancer nanomedicine and the approaches to master them. This oration draws mainly from my lab's journey to explore three transformative approaches to master complexities in the field: (1) leveraging text mining to construct dynamic knowledge bases for hypothesis generation in cell-specific drug delivery, (2) introducing the concept of meta-synergy to further optimize and classify multi-drug combinations across dimensions such as chemical loading, pharmacodynamics, and pharmacokinetics (3) utilizing automation to accelerate nanoparticle discovery with advanced screening methodologies such as aggregation-induced emission (AIE). I argue that by embracing complexity in nanomedicine, we can manifest new therapeutic possibilities, paving the way for more effective, precise, and adaptive treatment strategies.

Computer-Aided Design of Self-Assembled Nanoparticles to Enhance Cancer Chemoimmunotherapy via Dual-Modulation Strategy

The rational design of self-assembled compounds is crucial for the highly efficient development of carrier-free nanomedicines. Herein, based on computer-aided strategies, important physicochemical properties are identified to guide the rational design of self-assembled compounds. Then, the pharmacophore hybridization strategy is used to design self-assemble nanoparticles by preparing new chemical structures by combining pharmacophore groups of different bioactive compounds. Hydroxychloroquine is grafted with the lipophilic vitamin E succinate and then co-assembled with bortezomib to fabricate the nanoparticle. The nanoparticle can reduce M2-type tumor-associated macrophages (TAMs) through lysosomal alkalization and induce immunogenic cell death (ICD) and nuclear factor-κB (NF-κB) inhibition in tumor cells. In mouse models, the nanoparticles induce decreased levels of M2-type TAMs, regulatory T cells, and transforming growth factor-β (TGF-β), and increase the proportion of cytotoxicity T lymphocytes. Additionally, the nanoparticles reduce the secretion of Interleukin-6 (IL-6) by inhibiting NF-κB and enhance the programmed death ligand-1 (PD-L1) checkpoint blockade therapy. The pharmacophore hybridization-derived nanoparticle provides a dual-modulation strategy to reprogram the tumor microenvironment, which will efficiently enhance the chemoimmunotherapy against triple-negative breast cancer.

Machine Learning-Driven Prediction, Preparation, and Evaluation of Functional Nanomedicines Via Drug-Drug Self-Assembly

Small molecules as nanomedicine carriers offer advantages in drug loading and preparation. Selecting effective small molecules for stable nanomedicines is challenging. This study used artificial intelligence (AI) to screen drug combinations for self-assembling nanomedicines, employing physiochemical parameters to predict formation via machine learning. Non-Steroidal Anti-Inflammatory Drugs (NSAIDs) are identified as effective carriers for antineoplastic drugs, with high drug loading. Nanomedicines, PEG-coated indomethacin/paclitaxel nanomedicine (PiPTX), and laminarin-modified indomethacin/paclitaxel nanomedicine (LiDOX), are developed with extended circulation and active targeting functions. Indomethacin/paclitaxel nanomedicine iDOX exhibits pH-responsive drug release in the tumor microenvironment. These nanomedicines enhance anti-tumor effects and reduce side effects, offering a rapid approach to clinical nanomedicine development.

A Library of Polyphenol-Amino Acid Condensates for High-Throughput Continuous Flow Production of Nanomedicines with Ultra-High Drug Loading

Synthesizing high drug-loading nanomedicines remains a formidable challenge, and achieving universally applicable, continuous, large-scale engineered production of such nanomedicines presents even greater difficulties. This study presents a scalable library of polyphenol-amino acid condensates. By selecting amino acids, the library enables precise customization of key properties, such as carrier capacity, bioactivity, and other critical attributes, offering a versatile range of options for various application scenarios. Leveraging the properties of solvent-mediated disassembly and reassembly of condensates achieved an ultra-high drug loading of 86% for paclitaxel. For a range of poorly soluble molecules, the drug loading capacity exceeded 50%, indicating broad applicability. Furthermore, employing a continuous microfluidic device, the production rate can reach 5 mL min-1 (36 g per day), with the nanoparticle size precisely tunable and a polydispersity index (PDI) below 0.2. The polyphenol-based carrier demonstrates efficient cellular uptake and, in three distinct animal models, has been shown to enhance the therapeutic efficacy of paclitaxel without significant side effects. This study presents a streamlined, efficient, and scalable approach using microfluidics to produce nanomedicines with ultra-high drug loading, offering a promising strategy for the nanoformulation of poorly soluble drugs.

Artificial intelligence in dentistry and dental biomaterials

Artificial intelligence (AI) technology is being used in various fields and its use is increasingly expanding in dentistry. The key aspects of AI include machine learning (ML), deep learning (DL), and neural networks (NNs). The aim of this review is to present an overview of AI, its various aspects, and its application in biomedicine, dentistry, and dental biomaterials focusing on restorative dentistry and prosthodontics. AI-based systems can be a complementary tool in diagnosis and treatment planning, result prediction, and patient-centered care. AI software can be used to detect restorations, prosthetic crowns, periodontal bone loss, and root canal segmentation from the periapical radiographs. The integration of AI, digital imaging, and 3D printing can provide more precise, durable, and patient-oriented outcomes. AI can be also used for the automatic segmentation of panoramic radiographs showing normal anatomy of the oral and maxillofacial area. Recent advancement in AI in medical and dental sciences includes multimodal deep learning fusion, speech data detection, and neuromorphic computing. Hence, AI has helped dentists in diagnosis, planning, and aid in providing high-quality dental treatments in less time.

Molecular Design of Phthalocyanine-Based Drug Coassembly with Tailored Function

Coassemblies with tailored functions, such as drug loading, tissue targeting and releasing, therapeutic and/or imaging purposes, and so on, have been widely studied and applied in biomedicine. De novo design of these coassemblies hinges on an integrated approach involving synergy between various design strategies, ranging from structure screening of combinations of "phthalocyanine-chemotherapeutic drug" molecules for molecular scaffolds, exploration of related fabrication principles to verification of intended activity of specific designs. Here, we propose an integrated approach combining computation and experiments to design from scratch coassembled nanoparticles. This nanocoassembly, termed NanoPC here, consists of phthalocyanine-based scaffolds hosting chemotherapeutic drugs, aimed at hypersensitive chemotherapy guided by photoimaging for targeting tumors. Our design starts from the selection of phthalocyanine derivatives that are not aggregation-prone, followed by computational screening of coassembled molecules covering various categories of chemotherapy drugs. To facilitate an efficient and accurate assessment of coassembly capabilities, we utilize small systems as surrogates to enable free-energy calculations at all-atom levels facilitated with enhanced sampling and statistical mechanics for efficient and accurate evaluation of coassembly ability. The final top NanoPC candidate, comprised of phthalocyanine PcL and cytarabine (CYT), can greatly increase the fluorescence intensity ratio of tumor/liver by 21.5 times and achieve higher antitumor efficiency in a pH-dependent manner. Therefore, the designing approach proposed here has a potential pattern, which can provide ideas and references for the design and development of coassembled nanodrugs with tailored functions and applications in biomedicine.

Artificial intelligence-guided design of lipid nanoparticles for pulmonary gene therapy

Ionizable lipids are a key component of lipid nanoparticles, the leading nonviral messenger RNA delivery technology. Here, to advance the identification of ionizable lipids beyond current methods, which rely on experimental screening and/or rational design, we introduce lipid optimization using neural networks, a deep-learning strategy for ionizable lipid design. We created a dataset of >9,000 lipid nanoparticle activity measurements and used it to train a directed message-passing neural network for prediction of nucleic acid delivery with diverse lipid structures. Lipid optimization using neural networks predicted RNA delivery in vitro and in vivo and extrapolated to structures divergent from the training set. We evaluated 1.6 million lipids in silico and identified two structures, FO-32 and FO-35, with local mRNA delivery to the mouse muscle and nasal mucosa. FO-32 matched the state of the art for nebulized mRNA delivery to the mouse lung, and both FO-32 and FO-35 efficiently delivered mRNA to ferret lungs. Overall, this work shows the utility of deep learning for improving nanoparticle delivery.

Designing nanotheranostics with machine learning

The inherent limits of traditional diagnoses and therapies have driven the development and application of emerging nanotechnologies for more effective and safer management of diseases, herein referred to as 'nanotheranostics'. Although many important technological successes have been achieved in this field, widespread adoption of nanotheranostics as a new paradigm is hindered by specific obstacles, including time-consuming synthesis of nanoparticles, incomplete understanding of nano-bio interactions, and challenges regarding chemistry, manufacturing and the controls required for clinical translation and commercialization. As a key branch of artificial intelligence, machine learning (ML) provides a set of tools capable of performing time-consuming and result-perception tasks, thus offering unique opportunities for nanotheranostics. This Review summarizes the progress and challenges in this emerging field of ML-aided nanotheranostics, and discusses the opportunities in developing next-generation nanotheranostics with reliable datasets and advanced ML models to offer better clinical benefits to patients.

The Role of Artificial Intelligence and Machine Learning in Accelerating the Discovery and Development of Nanomedicine

The unique potential of nanomedicine to address challenging health issues is rapidly advancing the field, leading to the generation of more effective products. However, these complex systems often pose several challenges with respect to their design for specific functionality, scalable manufacturing, characterization, quality control, and clinical translation. In this regard, the application of artificial intelligence (AI) and machine learning (ML) approaches can enable faster and more accurate data assessment, identifying trends and predicting outcomes, leading to efficient nanomedicine product development. This perspective paper discusses the potential of AI and ML in nanomedicine product development with a focus on their applications in discovery, assessment, manufacturing, and clinical trials. The potential limitations of AI and ML approaches in nanomedicine product development are also covered.

Rational strategies for improving the efficiency of design and discovery of nanomedicines

The rise of rational strategies in nanomedicine development, such as high-throughput methods and computer-aided techniques, has led to a shift in the design and discovery patterns of nanomedicines from a trial-and-error mode to a rational mode. This transition facilitates the enhancement of efficiency in the preclinical discovery pipeline of nanomaterials, particularly in improving the hit rate of nanomaterials and the optimization efficiency of promising candidates. Herein, we describe a directed evolution mode of nanomedicines driven by data to accelerate the discovery of nanomaterials with high delivery efficiency. Computer-aided design strategies are introduced in detail as one of the cutting-edge directions for the development of nanomedicines. Ultimately, we look forward to expanding the tools for the rational design and discovery of nanomaterials using multidisciplinary approaches. Rational design strategies may potentially boost the delivery efficiency of next-generation nanomedicines.

Replenishing Cation-π Interactions for the Fabrication of Mesoporous Levodopa Nanoformulations for Parkinson Remission

Directly assembling drugs into mesoporous nanoformulations will be greatly favored due to the combination of enhanced drug delivery efficiency and mesostructure-enabled nanobio interactions. However, such an approach is hindered due to the lack of understanding of polymer nanoparticles' formation mechanism, especially the relationship between polymerization, self-assembly, and the nucleation process. Here, by investigating the levodopa and dopamine polymerization process, we identify π-cation interaction as pivotal in the self-assembly and nucleation control of dopa molecules. Thus, through manipulation of the π-cation interaction, we present the direct assembly of a commercial drug, levodopa, into mesoporous nanoformulations. The synthesized nanospheres, approximately 200 nm in diameter, exhibit uniform mesopores of around 8 nm. These nanoformulations, abundant in mesopores, enhance chiral phenylalanine interaction with α-synuclein (Syn), curbing aggregation, safeguarding neurons, and alleviating Parkinson's pathology. When combating α-synuclein, the nanoformulation achieved ∼100% inhibition of protein aggregation and sustained neuron viability up to 300%. We believe that this study may advance mesoscale self-assembly knowledge, guiding future nanopharmaceutical developments.

Integrated computational approaches for advancing antimicrobial peptide development

The increasing prevalence of antimicrobial resistance has intensified the need for novel antimicrobial drugs. Antimicrobial peptides (AMPs) are promising alternative antibiotics due to their broad-spectrum activity and slower resistance development. However, the time-consuming, costly development and challenge of systematic optimization limit their translation into the clinic. Recently, integrating computational methods have led to breakthroughs in the precise design and optimization of AMPs, reduced resource consumption, and accelerated AMP development process. We highlight the application of these integrated approaches in AMP molecule discovery, optimization, and delivery and demonstrate the synergy of these strategies to fuel AMP development.

Transformative Impact of Nanocarrier‐Mediated Drug Delivery: Overcoming Biological Barriers and Expanding Therapeutic Horizons

Advancing therapeutic progress is centered on developing drug delivery systems (DDS) that control therapeutic molecule release, ensuring precise targeting and optimal concentrations. Targeted DDS enhances treatment efficacy and minimizes off‐target effects, but struggles with drug degradation. Over the last three decades, nanopharmaceuticals have evolved from laboratory concepts into clinical products, highlighting the profound impact of nanotechnology in medicine. Despite advancements, the effective delivery of therapeutics remains challenging because of biological barriers. Nanocarriers offer a solution with a small size, high surface‐to‐volume ratios, and customizable properties. These systems address physiological and biological challenges, such as shear stress, protein adsorption, and quick clearance. They allow targeted delivery to specific tissues, improve treatment outcomes, and reduce adverse effects. Nanocarriers exhibit controlled release, decreased degradation, and enhanced efficacy. Their size facilitates cell membrane penetration and intracellular delivery. Surface modifications increase affinity for specific cell types, allowing precise treatment delivery. This study also elucidates the potential integration of artificial intelligence with nanoscience to innovate future nanocarrier systems.

Develop Targeted Protein Drug Carriers through a High-Throughput Screening Platform and Rational Design

Protein-based drugs offer advantages, such as high specificity, low toxicity, and minimal side effects compared to small molecule drugs. However, delivery of proteins to target tissues or cells remains challenging due to the instability, diverse structures, charges, and molecular weights of proteins. Polymers have emerged as a leading choice for designing effective protein delivery systems, but identifying a suitable polymer for a given protein is complicated by the complexity of both proteins and polymers. To address this challenge, a fluorescence-based high-throughput screening platform called ProMatch to efficiently collect data on protein-polymer interactions, followed by in vivo and in vitro experiments with rational design is developed. Using this approach to streamline polymer selection for targeted protein delivery, candidate polymers from commercially available options are identified and a polyhexamethylene biguanide (PHMB)-based system for delivering proteins to white adipose tissue as a treatment for obesity is developed. A branched polyethylenimine (bPEI)-based system for neuron-specific protein delivery to stimulate optic nerve regeneration is also developed. The high-throughput screening methodology expedites identification of promising polymer candidates for tissue-specific protein delivery systems, thereby providing a platform to develop innovative protein-based therapeutics.

Intranasal Carrier-Free Nanomodulator Addresses Both Symptomatology and Etiology of Alzheimer's Disease by Restoring Neuron Plasticity and Reprogramming Lesion Microenvironment

The unsatisfactory treatment outcome of Alzheimer's disease (AD) can be attributed to two primary factors, the intricate pathogenic mechanisms leading to restricted treatment effectiveness against single targets and the hindered drug accumulation in brain due to blood-brain barrier obstruction. Therefore, we developed a carrier-free nanomodulator (NanoDS) through the self-assembly of donepezil and simvastatin for direct nose-to-brain delivery. This approach facilitated a rapid and efficient traversal through the nasal epithelial barrier, enabling subsequent drug release and achieving multiple therapeutic effects. Among them, donepezil effectively ameliorated the symptoms of AD and restored synaptic plasticity. Simvastatin exerted a neurotrophic effect and facilitated the clearance of amyloid-β aggregation. At the same time, NanoDS demonstrated effective anti-inflammatory and antioxidative stress effects. This therapy for AD is approached from both symptomatic and etiological perspectives. In the treatment of FAD4T transgenic mice, it highly improved spatial memory impairment and cognitive deficits while restoring the homeostasis of brain microenvironment. Collectively, our study presented a paradigm for both achieving efficient brain delivery and offering pleiotropic therapies for AD.

Two-dimensional coordination risedronate-manganese nanobelts as adjuvant for cancer radiotherapy and immunotherapy

The irradiated tumor itself represents an opportunity to establish endogenous in situ vaccines. However, such in situ cancer vaccination (ISCV) triggered by radiation therapy (RT) alone is very weak and hardly elicits systemic anticancer immunity. In this study, we develop two-dimensional risedronate-manganese nanobelts (RMn-NBs) as an adjuvant for RT to address this issue. RMn-NBs exhibit good T2 magnetic resonance imaging performance and enhanced Fenton-like catalytic activity, which induces immunogenic cell death. RMn-NBs can inhibit the HIF-1α/VEGF axis to empower RT and synchronously activate the cGAS/STING pathway for promoting the secretion of type I interferon, thereby boosting RT-triggered ISCV and immune checkpoint blockade therapy against primary and metastatic tumors. RMn-NBs as a nano-adjuvant for RT show good biocompatibility and therapeutic efficacy, presenting a promising prospect for cancer radiotherapy and immunotherapy.

Mapping Biomaterial Complexity by Machine Learning

Biomaterials often have subtle properties that ultimately drive their bespoke performance. Given this nuanced structure-function behavior, the standard scientific approach of one experiment at a time or design of experiment methods is largely inefficient for the discovery of complex biomaterials. More recently, high-throughput experimentation coupled with machine learning methods has matured beyond expert users allowing scientists and engineers from diverse backgrounds to access these powerful data science tools. As a result, we now have the opportunity to strategically utilize all available data from high-throughput experiments to train efficacious models and map the structure-function behavior of biomaterials for their discovery. Herein, we discuss this necessary shift to data-driven determination of structure-function properties of biomaterials as we highlight how machine learning is leveraged in identifying physicochemical cues for biomaterials in tissue engineering, gene delivery, drug delivery, protein stabilization, and antifouling materials. We also discuss data-mining approaches that are coupled with machine learning to map biomaterial functions that reduce the load on experimental approaches for faster biomaterial discovery. Ultimately, harnessing the prowess of machine learning will lead to accelerated discovery and development of optimal biomaterial designs.

Computer-aided nanodrug discovery: recent progress and future prospects

Nanodrugs, which utilise nanomaterials in disease prevention and therapy, have attracted considerable interest since their initial conceptualisation in the 1990s. Substantial efforts have been made to develop nanodrugs for overcoming the limitations of conventional drugs, such as low targeting efficacy, high dosage and toxicity, and potential drug resistance. Despite the significant progress that has been made in nanodrug discovery, the precise design or screening of nanomaterials with desired biomedical functions prior to experimentation remains a significant challenge. This is particularly the case with regard to personalised precision nanodrugs, which require the simultaneous optimisation of the structures, compositions, and surface functionalities of nanodrugs. The development of powerful computer clusters and algorithms has made it possible to overcome this challenge through in silico methods, which provide a comprehensive understanding of the medical functions of nanodrugs in relation to their physicochemical properties. In addition, machine learning techniques have been widely employed in nanodrug research, significantly accelerating the understanding of bio-nano interactions and the development of nanodrugs. This review will present a summary of the computational advances in nanodrug discovery, focusing on the understanding of how the key interfacial interactions, namely, surface adsorption, supramolecular recognition, surface catalysis, and chemical conversion, affect the therapeutic efficacy of nanodrugs. Furthermore, this review will discuss the challenges and opportunities in computer-aided nanodrug discovery, with particular emphasis on the integrated "computation + machine learning + experimentation" strategy that can potentially accelerate the discovery of precision nanodrugs.

A nanocarbon-enabled hybridization strategy to construct pharmacologically cooperative therapeutics for augmented anticancer efficacy

The drug design principles are of great value in developing nanomedicines with favorable functionalities. Herein we propose a nanocarbon-enabled hybridization strategy to construct a pharmacologically cooperative nanodrug for improved cancer therapy in the light of pharmacophore hybridization in medicinal chemistry and the synthetic principles of nanocarbons. An antioxidant defense pharmacological inhibitor and a co-nucleation precursor are structurally hybridized into nanodrugs (SCACDs) via forming carbon quantum dots. These SCACDs elicit dual enhanced bioactivities, including superior sonocatalytic activity that arose from the appropriate band structure of the pharmacophoric carbon cores, and more than an order of magnitude higher antioxidant defense inhibitory activity than the pharmacological inhibitor via conveying the bioactive pharmacophores from the molecular level to nanoscale. In vivo, SCACDs possess a long body retention and desirable biodistribution to eliminate melanoma cells at a very low injection dose. The present study provides a viable yet effective strategy for the development of pharmacologically cooperative nanodrugs to achieve remarkably improved therapeutic efficacy.

AI in cellular engineering and reprogramming

During the last decade, artificial intelligence (AI) has increasingly been applied in biophysics and related fields, including cellular engineering and reprogramming, offering novel approaches to understand, manipulate, and control cellular function. The potential of AI lies in its ability to analyze complex datasets and generate predictive models. AI algorithms can process large amounts of data from single-cell genomics and multiomic technologies, allowing researchers to gain mechanistic insights into the control of cell identity and function. By integrating and interpreting these complex datasets, AI can help identify key molecular events and regulatory pathways involved in cellular reprogramming. This knowledge can inform the design of precision engineering strategies, such as the development of new transcription factor and signaling molecule cocktails, to manipulate cell identity and drive authentic cell fate across lineage boundaries. Furthermore, when used in combination with computational methods, AI can accelerate and improve the analysis and understanding of the intricate relationships between genes, proteins, and cellular processes. In this review article, we explore the current state of AI applications in biophysics with a specific focus on cellular engineering and reprogramming. Then, we showcase a couple of recent applications where we combined machine learning with experimental and computational techniques. Finally, we briefly discuss the challenges and prospects of AI in cellular engineering and reprogramming, emphasizing the potential of these technologies to revolutionize our ability to engineer cells for a variety of applications, from disease modeling and drug discovery to regenerative medicine and biomanufacturing.

Machine learning in drug delivery

For decades, drug delivery scientists have been performing trial-and-error experimentation to manually sample parameter spaces and optimize release profiles through rational design. To enable this approach, scientists spend much of their career learning nuanced drug-material interactions that drive system behavior. In relatively simple systems, rational design criteria allow us to fine tune release profiles and enable efficacious therapies. However, as materials and drugs become increasingly sophisticated and their interactions have non-linear and compounding effects, the field is suffering the Curse of Dimensionality which prevents us from comprehending complex structure-function relationships. In the past, we have embraced this complexity by implementing high-throughput screens to increase the probability of finding ideal compositions. However, this brute force method was inefficient and led many to abandon these fishing expeditions. Fortunately, methods in data science including artificial intelligence / machine learning (AI/ML) are providing ideal analytical tools to model this complex data and ascertain quantitative structure-function relationships. In this Oration, I speak to the potential value of data science in drug delivery with particular focus on polymeric delivery systems. Here, I do not suggest that AI/ML will simply replace mechanistic understanding of complex systems. Rather, I propose that AI/ML should be yet another useful tool in the lab to navigate complex parameter spaces. The recent hype around AI/ML is breathtaking and potentially over inflated, but the value of these methods is poised to revolutionize how we perform science. Therefore, I encourage readers to consider adopting these skills and applying data science methods to their own problems. If done successfully, I believe we will all realize a paradigm shift in our approach to drug delivery.

AI-powered omics-based drug pair discovery for pyroptosis therapy targeting triple-negative breast cancer

Due to low success rates and long cycles of traditional drug development, the clinical tendency is to apply omics techniques to reveal patient-level disease characteristics and individualized responses to treatment. However, the heterogeneous form of data and uneven distribution of targets make drug discovery and precision medicine a non-trivial task. This study takes pyroptosis therapy for triple-negative breast cancer (TNBC) as a paradigm and uses data mining of a large TNBC cohort and drug databases to establish a biofactor-regulated neural network for rapidly screening and optimizing compound pyroptosis drug pairs. Subsequently, biomimetic nanococrystals are prepared using the preferred combination of mitoxantrone and gambogic acid for rational drug delivery. The unique mechanism of obtained nanococrystals regulating pyroptosis genes through ribosomal stress and triggering pyroptosis cascade immune effects are revealed in TNBC models. In this work, a target omics-based intelligent compound drug discovery framework explores an innovative drug development paradigm, which repurposes existing drugs and enables precise treatment of refractory diseases.

Network pharmacology combined with affinity ultrafiltration to elucidate the potential compounds of Shaoyao Gancao Fuzi Decoction for the treatment of rheumatoid arthritis

ETHNOPHARMACOLOGICAL RELEVANCE

Shaoyao Gancao Fuzi Decoction (SGFD), has been employed for thousands of years in the treatment of rheumatoid arthritis (RA) with remarkable clinical efficacy. However, the material basis underlying the effectiveness of SGFD still remains unclear.

AIM OF THE REVIEW

This study aims to elucidate the material basis of SGFD through the application of network pharmacology and biological affinity ultrafiltration.

RESULTS

UPLC-Q-TOF-MS/MS was employed to characterize the components in SGFD, the identified 145 chemical components were mainly categorized into alkaloids, flavonoids, triterpenoids, and monoterpenoids according to the structures. Network pharmacology method was utilized to identify potential targets and signaling pathways of SGFD in the RA treatment, and the anti-inflammatory and anti-RA effects of SGFD were validated through in vivo and in vitro experiments. Moreover, as the significant node in the pharmacology network, TNF-α, a classical therapeutic target in RA, was subsequent employed to screen the interacting compounds in SGFD via affinity ultrafiltration screening method, 6 active molecules (i.e.,glycyrrhizic acid, paeoniflorin, formononetin, isoliquiritigenin, benzoyl mesaconitine, and glycyrrhetinic acid) were exhibited significant interactions. Finally, the significant anti-inflammatory and anti-TNF-α effects of these compounds were validated at the cellular level.

CONCLUSIONS

In conclusion, this study comprehensively elucidates the pharmacodynamic material basis of SGFD, offering a practical reference model for the systematic investigation of traditional Chinese medicine formulas.

Scientific Discovery Framework Accelerating Advanced Polymeric Materials Design

Organic polymer materials, as the most abundantly produced materials, possess a flammable nature, making them potential hazards to human casualties and property losses. Target polymer design is still hindered due to the lack of a scientific foundation. Herein, we present a robust, generalizable, yet intelligent polymer discovery framework, which synergizes diverse capabilities, including the in situ burning analyzer, virtual reaction generator, and material genomic model, to achieve results that surpass the sum of individual parts. Notably, the high-throughput analyzer created for the first time, grounded in multiple spectroscopic principles, enables in situ capturing of massive combustion intermediates; then, the created realistic apparatus transforming to the virtual reaction generator acquires exponentially more intermediate information; further, the proposed feature engineering tool, which embedded both polymer hierarchical structures and massive intermediate data, develops the generalizable genomic model with excellent universality (adapting over 20 kinds of polymers) and high accuracy (88.8%), succeeding discovering series of novel polymers. This emerging approach addresses the target polymer design for flame-retardant application and underscores a pivotal role in accelerating polymeric materials discovery.

Data-driven learning of structure augments quantitative prediction of biological responses

Multi-factor screenings are commonly used in diverse applications in medicine and bioengineering, including optimizing combination drug treatments and microbiome engineering. Despite the advances in high-throughput technologies, large-scale experiments typically remain prohibitively expensive. Here we introduce a machine learning platform, structure-augmented regression (SAR), that exploits the intrinsic structure of each biological system to learn a high-accuracy model with minimal data requirement. Under different environmental perturbations, each biological system exhibits a unique, structured phenotypic response. This structure can be learned based on limited data and once learned, can constrain subsequent quantitative predictions. We demonstrate that SAR requires significantly fewer data comparing to other existing machine-learning methods to achieve a high prediction accuracy, first on simulated data, then on experimental data of various systems and input dimensions. We then show how a learned structure can guide effective design of new experiments. Our approach has implications for predictive control of biological systems and an integration of machine learning prediction and experimental design.

A large-scale machine learning analysis of inorganic nanoparticles in preclinical cancer research

Owing to their distinct physical and chemical properties, inorganic nanoparticles (NPs) have shown promising results in preclinical cancer therapy, but designing and engineering them for effective therapeutic purposes remains a challenge. Although a comprehensive database of inorganic NP research is not currently available, it is crucial for developing effective cancer therapies. In this context, machine learning (ML) has emerged as a transformative tool, but its adaptation to nanomedicine is hindered by inexistent or small datasets. Here we assembled a large database of inorganic NPs, comprising experimental datasets from 745 preclinical studies in cancer nanomedicine. Using descriptive statistics and explainable ML models we mined this database to gain knowledge of inorganic NP design patterns and inform future NP research for cancer treatment. Our analyses suggest that NP shape and therapy type are prominent features in determining in vivo efficacy, measured as a percentage of tumour reduction. Moreover, our database provides a large-scale open-access resource for discriminative ML that the broader nanotechnology community can utilize. Our work blueprints data mining for translational cancer research and offers evidence for standardizing NP reporting to accelerate and de-risk inorganic NP-based drug delivery, which may help to improve patient outcomes in clinical settings.

Step-By-Step Standardization of the Bottom-Up Semi-Automated Nanocrystallization of Pharmaceuticals: A Quality By Design and Design of Experiments Joint Approach

Nanosized drug crystals have been reported with enhanced apparent solubility, bioavailability, and therapeutic efficacy compared to microcrystal materials, which are not suitable for parenteral administration. However, nanocrystal design and development by bottom-up approaches are challenging, especially considering the non-standardized process parameters in the injection step. This work aims to present a systematic step-by-step approach through Quality-by-Design (QbD) and Design of Experiments (DoE) for synthesizing drug nanocrystals by a semi-automated nanoprecipitation method. Curcumin is used as a drug model due to its well-known poor water solubility (0.6 µg mL-1, 25 °C). Formal and informal risk assessment tools allow identifying the critical factors. A fractional factorial 24-1 screening design evaluates their impact on the average size and polydispersity of nanocrystals. The optimization of significant factors is done by a Central Composite Design. This response surface methodology supports the rational design of the nanocrystals, identifying and exploring the design space. The proposed joint approach leads to a reproducible, robust, and stable nanocrystalline preparation of 316 nm with a PdI of 0.217 in compliance with the quality profile. An orthogonal approach for particle size and polydispersity characterization allows discarding the formation of aggregates. Overall, the synergy between advanced data analysis and semi-automated standardized nanocrystallization of drugs is highlighted.

Intelligent Biomaterialomics: Molecular Design, Manufacturing, and Biomedical Applications

Materialomics integrates experiment, theory, and computation in a high-throughput manner, and has changed the paradigm for the research and development of new functional materials. Recently, with the rapid development of high-throughput characterization and machine-learning technologies, the establishment of biomaterialomics that tackles complex physiological behaviors has become accessible. Breakthroughs in the clinical translation of nanoparticle-based therapeutics and vaccines have been observed. Herein, recent advances in biomaterials, including polymers, lipid-like materials, and peptides/proteins, discovered through high-throughput screening or machine learning-assisted methods, are summarized. The molecular design of structure-diversified libraries; high-throughput characterization, screening, and preparation; and, their applications in drug delivery and clinical translation are discussed in detail. Furthermore, the prospects and main challenges in future biomaterialomics and high-throughput screening development are highlighted.

The landscape of small-molecule prodrugs

Prodrugs are derivatives with superior properties compared with the parent active pharmaceutical ingredient (API), which undergo biotransformation after administration to generate the API in situ. Although sharing this general characteristic, prodrugs encompass a wide range of different chemical structures, therapeutic indications and properties. Here we provide the first holistic analysis of the current landscape of approved prodrugs using cheminformatics and data science approaches to reveal trends in prodrug development. We highlight rationales that underlie prodrug design, their indications, mechanisms of API release, the chemistry of promoieties added to APIs to form prodrugs and the market impact of prodrugs. On the basis of this analysis, we discuss strengths and limitations of current prodrug approaches and suggest areas for future development.

Expanding the Horizons of Machine Learning in Nanomaterials to Chiral Nanostructures

Machine learning holds significant research potential in the field of nanotechnology, enabling nanomaterial structure and property predictions, facilitating materials design and discovery, and reducing the need for time-consuming and labor-intensive experiments and simulations. In contrast to their achiral counterparts, the application of machine learning for chiral nanomaterials is still in its infancy, with a limited number of publications to date. This is despite the great potential of machine learning to advance the development of new sustainable chiral materials with high values of optical activity, circularly polarized luminescence, and enantioselectivity, as well as for the analysis of structural chirality by electron microscopy. In this review, an analysis of machine learning methods used for studying achiral nanomaterials is provided, subsequently offering guidance on adapting and extending this work to chiral nanomaterials. An overview of chiral nanomaterials within the framework of synthesis-structure-property-application relationships is presented and insights on how to leverage machine learning for the study of these highly complex relationships are provided. Some key recent publications are reviewed and discussed on the application of machine learning for chiral nanomaterials. Finally, the review captures the key achievements, ongoing challenges, and the prospective outlook for this very important research field.

Combinatorial discovery of antibacterials via a feature-fusion based machine learning workflow

The discovery of new antibacterials within the vast chemical space is crucial in combating drug-resistant bacteria such as methicillin-resistant Staphylococcus aureus (MRSA). However, the traditional approach of screening the entire chemical library in an ergodic manner can be laborious and time-consuming. Machine learning-assisted screening of antibacterials alleviates the exploration effort but suffers from the lack of reliable and related datasets. To address these challenges, we devised a combinatorial library comprising over 110 000 candidates based on the Ugi reaction. A focused library was subsequently generated through uniform sampling of the entire library to narrow down the preliminary screening scale. A novel feature-fusion architecture called the latent space constraint neural network was developed which incorporated both fingerprint and physicochemical molecular descriptors to predict the antibacterial properties. This integration allowed the model to leverage the complementary information provided by these descriptors and improve the accuracy of predictions. Three lead compounds that demonstrated excellent efficacy against MRSA while alleviating drug resistance were identified. This workflow highlights the integration of machine learning with the combinatorial chemical library to expedite high-quality data collection and extensive data mining for antibacterial screening.

AI energized hydrogel design, optimization and application in biomedicine

Traditional hydrogel design and optimization methods usually rely on repeated experiments, which is time-consuming and expensive, resulting in a slow-moving of advanced hydrogel development. With the rapid development of artificial intelligence (AI) technology and increasing material data, AI-energized design and optimization of hydrogels for biomedical applications has emerged as a revolutionary breakthrough in materials science. This review begins by outlining the history of AI and the potential advantages of using AI in the design and optimization of hydrogels, such as prediction and optimization of properties, multi-attribute optimization, high-throughput screening, automated material discovery, optimizing experimental design, and etc. Then, we focus on the various applications of hydrogels supported by AI technology in biomedicine, including drug delivery, bio-inks for advanced manufacturing, tissue repair, and biosensors, so as to provide a clear and comprehensive understanding of researchers in this field. Finally, we discuss the future directions and prospects, and provide a new perspective for the research and development of novel hydrogel materials for biomedical applications.

Strategies for non-viral vectors targeting organs beyond the liver

In recent years, nanoparticles have evolved to a clinical modality to deliver diverse nucleic acids. Rising interest in nanomedicines comes from proven safety and efficacy profiles established by continuous efforts to optimize physicochemical properties and endosomal escape. However, despite their transformative impact on the pharmaceutical industry, the clinical use of non-viral nucleic acid delivery is limited to hepatic diseases and vaccines due to liver accumulation. Overcoming liver tropism of nanoparticles is vital to meet clinical needs in other organs. Understanding the anatomical structure and physiological features of various organs would help to identify potential strategies for fine-tuning nanoparticle characteristics. In this Review, we discuss the source of liver tropism of non-viral vectors, present a brief overview of biological structure, processes and barriers in select organs, highlight approaches available to reach non-liver targets, and discuss techniques to accelerate the discovery of non-hepatic therapies.

New concepts drive the development of delivery tools for sustainable treatment of diabetic complications

Diabetic complications, especially diabetic retinopathy, diabetic nephropathy and painful diabetic neuropathy, account for a large portion of patients with diabetes and display rising global prevalence. They are the leading causes of blindness, kidney failure and hypersensitivity to pain caused by diabetes. Current approved therapeutics against the diabetic complications are few and exhibit limited efficacy. The enhanced cell-specificity, stability, biocompatibility, and loading capacity of drugs are essential for the mitigation of diabetic complications. In the article, we have critically discussed the recent studies over the past two years in material sciences and biochemistry. The insightful concepts in these studies drive the development of novel nanoparticles and mesenchymal stem cells-derived extracellular vesicles to meet the need for treatment of diabetic complications. Their underlying biochemical principles, advantages and limitations have been in-depth analyzed. The nanoparticles discussed in the article include double-headed nanodelivery system, nanozyme, ESC-HCM-B system, soft polymer nanostars, tetrahedral DNA nanostructures and hydrogels. They ameliorate the diabetic complication through attenuation of inflammation, apoptosis and restoration of metabolic homeostasis. Moreover, mesenchymal stem cell-derived extracellular vesicles efficiently deliver therapeutic proteins to the retinal cells to suppress the angiogenesis, inflammation, apoptosis and oxidative stress to reverse diabetic retinopathy. Collectively, we provide a critical discussion on the concept, mechanism and therapeutic applicability of new delivery tools to treat these three devastating diabetic complications.

Applying artificial intelligence and computational modeling to nanomedicine

Achieving specific and targeted delivery of nanomedicines to diseased tissues is a major challenge. This is because the process of designing, formulating, testing, and selecting a nanoparticle delivery vehicle for a specific disease target is governed by complex multivariate interactions. Computational modeling and artificial intelligence are well-suited for analyzing and modeling large multivariate datasets in short periods of time. Computational approaches can be applied to help design nanomedicine formulations, interpret nanoparticle-biological interactions, and create models from high-throughput screening techniques to improve the selection of the ideal nanoparticle carrier. In the future, many steps in the nanomedicine development process will be done computationally, reducing the number of experiments and time needed to select the ideal nanomedicine formulation.

Self-Assembled Nanotubes Based on Chiral H8-BINOL Modified with 1,2,3-Triazole to Recognize Bi3+ Efficiently by ICT Mechanism

A novel fluorescent "off" probe R-β-D-1 containing a 1,2,3-triazole moiety was obtained by the Click reaction with azidoglucose using H8-BINOL as a substrate, and the structure was characterized by 1H NMR and 13C NMR and ESI-MS analysis. The fluorescence properties of R-β-D-1 in methanol were investigated, and it was found that R-β-D-1 could be selectively fluorescently quenched by Bi3+ in the recognition of 19 metal ions and basic cations. The recognition process of Bi3+ by R-β-D-1 was also investigated by fluorescence spectroscopy, SEM, AFM, etc. The complex pattern of R-β-D-1 with Bi3+ was determined by Job's curve as 1 + 1, and the binding constant Ka of R-β-D-1 and Bi3+ was valued by the Benesi-Hildebrand equation as 1.01 × 104 M-1, indicating that the binding force of R-β-D-1 and Bi3+ was medium. The lowest detection limit (LOD) of the self-assembled H8-BINOL derivative for Bi3+ was up to 0.065 µM. The mechanism for the recognition of Bi3+ by the sensor R-β-D-1 may be the intramolecular charge transfer effect (ICT), which was attributed to the fact that the N-3 of the triazole readily serves as an electron acceptor while the incorporation of Bi3+ serves as an electron donor, and the two readily undergo coordination leading to the quenching of fluorescence. The recognition mechanism and recognition site could be verified by DFT calculation and CDD (Charge Density Difference).

Transferosomes as a transdermal drug delivery system: Dermal kinetics and recent developments

The development of innovative approaches to deliver medications has been growing now for the last few decades and generates a growing interest in the dermatopharmaceutical field. Transdermal drug delivery in particular, remains an attractive alternative route for many therapeutics. However, due to the limitations posed by the barrier properties of the stratum corneum, the delivery of many pharmaceutical dosage forms remains a challenge. Most successful therapies using the transdermal route have been ones containing smaller lipophilic molecules with molecular weights of a few hundred Daltons. To overcome these limitations of size and lipophilicity of the drugs, transferosomes have emerged as a successful tool for transdermal delivery of a variety of therapeutics including hydrophilic actives, larger molecules, peptides, proteins, and nucleic acids. Transferosomes exhibit a flexible structure and higher surface hydrophilicity which both play a critical role in the transport of drugs and other solutes using hydration gradients as a driving force to deliver the molecules into and across the skin. This results in enhanced overall permeation as well as controlled release of the drug in the skin layers. Additionally, the physical-chemical properties of the transferosomes provide increased stability by preventing degradation of the actives by oxidation, light, and temperature. Here, we present the history of transferosomes from solid lipid nanoparticles and liposomes, their physical-chemical properties, dermal kinetics, and their recent advances as marketed dosage forms. This article is categorized under: Biology-Inspired Nanomaterials > Lipid-Based Structures Therapeutic Approaches and Drug Discovery > Emerging Technologies.

The Potential of Congo Red Supplied Aggregates of Multitargeted Tyrosine Kinase Inhibitor (Sorafenib, BAY-43-9006) in Enhancing Therapeutic Impact on Bladder Cancer

Bladder cancer is a common malignancy associated with high recurrence rates and potential progression to invasive forms. Sorafenib, a multi-targeted tyrosine kinase inhibitor, has shown promise in anti-cancer therapy, but its cytotoxicity to normal cells and aggregation in solution limits its clinical application. To address these challenges, we investigated the formation of supramolecular aggregates of sorafenib with Congo red (CR), a bis-azo dye known for its supramolecular interaction. We analyzed different mole ratios of CR-sorafenib aggregates and evaluated their effects on bladder cancer cells of varying levels of malignancy. In addition, we also evaluated the effect of the test compounds on normal uroepithelial cells. Our results demonstrated that sorafenib inhibits the proliferation of bladder cancer cells and induces apoptosis in a dose-dependent manner. However, high concentrations of sorafenib also showed cytotoxicity to normal uroepithelial cells. In contrast, the CR-BAY aggregates exhibited reduced cytotoxicity to normal cells while maintaining anti-cancer activity. The aggregates inhibited cancer cell migration and invasion, suggesting their potential for metastasis prevention. Dynamic light scattering and UV-VIS measurements confirmed the formation of stable co-aggregates with distinctive spectral properties. These CR-sorafenib aggregates may provide a promising approach to targeted therapy with reduced cytotoxicity and improved stability for drug delivery in bladder cancer treatment. This work shows that the drug-excipient aggregates proposed and described so far, as Congo red-sorafenib, can be a real step forward in anti-cancer therapies.

Fragment-based drug nanoaggregation reveals drivers of self-assembly

Drug nanoaggregates are particles that can deleteriously cause false positive results during drug screening efforts, but alternatively, they may be used to improve pharmacokinetics when developed for drug delivery purposes. The structural features of molecules that drive nanoaggregate formation remain elusive, however, and the prediction of intracellular aggregation and rational design of nanoaggregate-based carriers are still challenging. We investigate nanoaggregate self-assembly mechanisms using small molecule fragments to identify the critical molecular forces that contribute to self-assembly. We find that aromatic groups and hydrogen bond acceptors/donors are essential for nanoaggregate formation, suggesting that both π-π stacking and hydrogen bonding are drivers of nanoaggregation. We apply structure-assembly-relationship analysis to the drug sorafenib and discover that nanoaggregate formation can be predicted entirely using drug fragment substructures. We also find that drug nanoaggregates are stabilized in an amorphous core-shell structure. These findings demonstrate that rational design can address intracellular aggregation and pharmacologic/delivery challenges in conventional and fragment-based drug development processes.

Application of ensemble machine learning approach to assess the factors affecting size and polydispersity index of liposomal nanoparticles

Liposome nanoparticles have emerged as promising drug delivery systems due to their unique properties. Assessing particle size and polydispersity index (PDI) is critical for evaluating the quality of these liposomal nanoparticles. However, optimizing these parameters in a laboratory setting is both costly and time-consuming. This study aimed to apply a machine learning technique to assess the impact of specific factors, including sonication time, extrusion temperature, and compositions, on the size and PDI of liposomal nanoparticles. Liposomal solutions were prepared and subjected to sonication with varying values for these parameters. Two compositions: (A) HSPC:DPPG:Chol:DSPE-mPEG2000 at 55:5:35:5 molar ratio and (B) HSPC:Chol:DSPE-mPEG2000 at 55:40:5 molar ratio, were made using remote loading method. Ensemble learning (EL), a machine learning technique, was employed using the Least-squares boosting (LSBoost) algorithm to accurately model the data. The dataset was randomly split into training and testing sets, with 70% allocated for training. The LSBoost algorithm achieved mean absolute errors of 1.652 and 0.0105 for modeling the size and PDI, respectively. Under conditions where the temperature was set at approximately 60 °C, our EL model predicted a minimum particle size of 116.53 nm for composition (A) with a sonication time of approximately 30 min. Similarly, for composition (B), the model predicted a minimum particle size of 129.97 nm with sonication times of approximately 30 or 55 min. In most instances, a PDI of less than 0.2 was achieved. These results highlight the significant impact of optimizing independent factors on the characteristics of liposomal nanoparticles and demonstrate the potential of EL as a decision support system for identifying the best liposomal formulation. We recommend further studies to explore the effects of other independent factors, such as lipid composition and surfactants, on liposomal nanoparticle characteristics.

STING agonist-boosted mRNA immunization via intelligent design of nanovaccines for enhancing cancer immunotherapy

Messenger RNA (mRNA) vaccine is revolutionizing the methodology of immunization in cancer. However, mRNA immunization is drastically limited by multistage biological barriers including poor lymphatic transport, rapid clearance, catalytic hydrolysis, insufficient cellular entry and endosome entrapment. Herein, we design a mRNA nanovaccine based on intelligent design to overcome these obstacles. Highly efficient nanovaccines are carried out with machine learning techniques from datasets of various nanocarriers, ensuring successful delivery of mRNA antigen and cyclic guanosine monophosphate-adenosine monophosphate (cGAMP) to targets. It activates stimulator of interferon genes (STING), promotes mRNA-encoded antigen presentation and boosts antitumour immunity in vivo, thus inhibiting tumour growth and ensuring long-term survival of tumour-bearing mice. This work provides a feasible and safe strategy to facilitate STING agonist-synergized mRNA immunization, with great translational potential for enhancing cancer immunotherapy.

Interpretable Molecular Property Predictions Using Marginalized Graph Kernels

Marginalized graph kernels have shown competitive performance in molecular machine learning tasks but currently lack measures of interpretability, which are important to improve trust in the models, detect biases, and inform molecular optimization campaigns. We here conceive and implement two interpretability measures for Gaussian process regression using a marginalized graph kernel (GPR-MGK) to quantify (1) the contribution of specific training data to the prediction and (2) the contribution of specific nodes of the graph to the prediction. We demonstrate the applicability of these interpretability measures for molecular property prediction. We compare GPR-MGK to graph neural networks on four logic and two real-world toxicology data sets and find that the atomic attribution of GPR-MGK generally outperforms the atomic attribution of graph neural networks. We also perform a detailed molecular attribution analysis using the FreeSolv data set, showing how molecules in the training set influence machine learning predictions and why Morgan fingerprints perform poorly on this data set. This is the first systematic examination of the interpretability of GPR-MGK and thereby is an important step in the further maturation of marginalized graph kernel methods for interpretable molecular predictions.

Prediction of cancer nanomedicines self-assembled from meta-synergistic drug pairs

Combination therapy is widely used in cancer medicine due to the benefits of drug synergy and the reduction of acquired resistance. To minimize emergent toxicities, nanomedicines containing drug combinations are being developed, and they have shown encouraging results. However, developing multi-drug loaded nanoparticles is highly complex and lacks predictability. Previously, it was shown that single drugs can self-assemble with near-infrared dye, IR783, to form cancer-targeted nanoparticles. A structure-based predictive model showed that only 4% of the drug space self-assembles with IR783. Here, we mapped the self-assembly outcomes of 77 small molecule drugs and drug pairs with IR783. We found that the small molecule drug space can be divided into five types, and type-1 drugs self-assemble with three out of four possible drug types that do not form stable nanoparticles. To predict the self-assembly outcome of any drug pair, we developed a machine learning model based on decision trees, which was trained and tested with F1-scores of 89.3% and 87.2%, respectively. We used literature text mining to capture drug pairs with biological synergy together with synergistic chemical self-assembly and generated a database with 1985 drug pairs for 70 cancers. We developed an online search tool to identify cancer-specific, meta-synergistic drug pairs (both chemical and biological synergism) and validated three different pairs in vitro. Lastly, we discovered a novel meta-synergistic pair, bortezomib-cabozantinib, which formed stable nanoparticles with improved biodistribution, efficacy, and reduced toxicity, even over single drugs, in an in vivo model of head and neck cancer.

Towards artificial intelligence-enabled extracellular vesicle precision drug delivery

Extracellular Vesicles (EVs), particularly exosomes, recently exploded into nanomedicine as an emerging drug delivery approach due to their superior biocompatibility, circulating stability, and bioavailability in vivo. However, EV heterogeneity makes molecular targeting precision a critical challenge. Deciphering key molecular drivers for controlling EV tissue targeting specificity is in great need. Artificial intelligence (AI) brings powerful prediction ability for guiding the rational design of engineered EVs in precision control for drug delivery. This review focuses on cutting-edge nano-delivery via integrating large-scale EV data with AI to develop AI-directed EV therapies and illuminate the clinical translation potential. We briefly review the current status of EVs in drug delivery, including the current frontier, limitations, and considerations to advance the field. Subsequently, we detail the future of AI in drug delivery and its impact on precision EV delivery. Our review discusses the current universal challenge of standardization and critical considerations when using AI combined with EVs for precision drug delivery. Finally, we will conclude this review with a perspective on future clinical translation led by a combined effort of AI and EV research.