Drug Discovery Pipeline Research Articles

Quantum mechanics/molecular mechanics (QM/MM) methods in drug design: a comprehensive review of development and applications

The integration of quantum mechanics (QM) and molecular mechanics (MM) methods, known as QM/MM, has emerged as a powerful computational approach in drug design. This hybrid technique combines the accuracy of QM calculations for the reactive region with the computational efficiency of MM methods for the surrounding environment. QM/MM methods have proven invaluable in studying chemical reactions, exploring enzyme mechanisms, and investigating ligand-protein interactions, all of which are crucial for rational drug design. This review provides a comprehensive overview of the QM/MM methodology, its theoretical foundations, and its applications in various aspects of drug design. We discussed the key components, such as QM and MM region partitioning, link atom schemes, and boundary treatments. Additionally, we highlight recent advancements and challenges in QM/MM methods, including polarizable force fields, implicit solvation models, and enhanced sampling techniques. Furthermore, we present illustrative examples showcasing the successful application of QM/MM methods in lead optimization, virtual screening, and the elucidation of biochemical mechanisms relevant to drug targets. Finally, we provide perspectives on future developments and the potential impact of QM/MM methods on the drug discovery pipeline.

Enhancing Ketoprofen Solubility: A Strategic Approach Using Solid Dispersion and Response Surface Methodology.

In the pharmaceutical sciences, the solubility profile of therapeutic molecules is crucial for identifying and formulating drugs and evaluating their quality across the drug discovery pipeline based on factors like oral bioavailability, metabolic transformation, biodistribution kinetics, and potential toxicological implications. The investigation aims to enhance the solubility parameters of ketoprofen (BCS-II class), which exhibits low solubility and high permeability. In this method, hydrotrope blends of aromatic sodium benzoate and electrolyte sodium acetate were employed to enhance the solubility parameter of ketoprofen. Several batches of solid dispersion of ketoprofen were made using a solvent evaporation method, and the response surface method 3² factorial design was used to find the best one. The optimised formulation, KSD9, underwent in-vitro drug dissolution, DSC, pXRD, and SEM studies. The optimized batch demonstrated substantial improvement in ketoprofen solubility, attributed to mixed hydrotropy. The results indicated that both solubility and %CDR improved when hydrotropes were employed, suggesting a direct proportionality between the rise in solubility and %CDR. Formulations KSD1-KSD9 exhibited solubility enhancements ranging from 2.23 to 5.77-fold, along with an elevation in %CDR from 72.28% to 94.76%. This implies that the %CDR was modulated by the hydrotropes, specifically influenced by the concentration levels of the independent variables. An increase in hydrotrope levels corresponded to an increase in %CDR. The positive coefficients in the quadratic equation for %CDR underscored the significant role of these independent variables in augmenting the in-vitro release of Ketoprofen. Similarly, during a comparative dissolution investigation, the optimized KSD9 formulation exhibited remarkable solubility and drug content compared to conventional Ketoprofen dispersible tablets. The synergistic effect of combining two hydrotropic agents significantly increased the solubility of ketoprofen by up to 58 times. The results indicated that the independent variables exerted a positive influence on solubility and %CDR. Furthermore, the responses were contingent on the specific hydrotropes selected, which functioned as the independent variables. Analyzing the r² and ANOVA results suggested that the dependent variables aligned well with the chosen model. Visual representations, such as the 3D response surface plot and contour plot, demonstrated the impact of each hydrotrope individually and when combined. Overall, employing hydrotropes led to improved solubility and %CDR, highlighting a direct proportionality between the rise in solubility and %CDR. Mixed hydrotropic lessens the toxicity associated with individual hydrotrope concentrations while also offering a sustainable and eco-friendly alternative. This study paves the way for future research aiming to improve the solubility of low- solubility drugs, broadening their clinical applications.

Machine Learning-Driven Data Valuation for Optimizing High-Throughput Screening Pipelines.

In the rapidly evolving field of drug discovery, high-throughput screening (HTS) is essential for identifying bioactive compounds. This study introduces a novel application of data valuation, a concept for evaluating the importance of data points based on their impact, to enhance drug discovery pipelines. Our approach improves active learning for compound library screening, robustly identifies true and false positives in HTS data, and identifies important inactive samples in an imbalanced HTS training, all while accounting for computational efficiency. We demonstrate that importance-based methods enable more effective batch screening, reducing the need for extensive HTS. Machine learning models accurately differentiate true biological activity from assay artifacts, streamlining the drug discovery process. Additionally, importance undersampling aids in HTS data set balancing, improving machine learning performance without omitting crucial inactive samples. These advancements could significantly enhance the efficiency and accuracy of drug development.

APPLICATION OF IN VITRO MODELS: INCORPORATING BLOOD BRAIN BARRIER MODELS WITH GLIOMA, AND THE USE OF MICROflUIDICS

Abstract AIMS The blood brain barrier (BBB) presents one of the main obstacles for the development of novel glioblastoma treatments. The various transporters and enzymes within the BBB reduce the disposition of novel therapies into the brain. There is a need for an in vitro BBB-glioma model with proven efficacy of predicting CNS delivery at earlier stages in the drug discovery pipeline. METHOD A BBB-glioma lab-on-a-chip model could enable rapid identification of novel therapies that can cross the BBB at earlier stages of the drug development process, before going to clinical trials. The application of microfluidic technology allows for the detection of soluble tissue-derived factors and the assessment of the levels of soluble markers released from malignant tissue. Adding ‘flow’ to the model, is more reflective of the dynamic flow present at the BBB. RESULTS Validation studies of the current BBB model and converting this model over to a lab-on-a-chip model have shown the presence of an endothelial cell barrier and the tight junction proteins Zo-1 and Claudin-1, through western blotting, immunocytochemistry, and permeability experiments using varying molecular weights of FitC-dextran. These validation studies have been completed using brain endothelial monolayers on polycarbonate transwell inserts, with the immortalised cell line HCMEC/d3, and the short-term culture cells HBMEC. Also, validation studies using a co-culture of endothelial cells, astrocytes and pericytes have begun, using the same techniques to show a barrier has formed, on both the current BBB model and the lab-on-a-chip model. CONCLUSION The aim of this study is to combine the use of an all-human BBB-glioma model with lab-on-a-chip technology to produce a high throughput model to test novel therapies at an earlier stage. The lab-on-a-chip model will be validated for the purpose of measuring permeability and drug response and to investigate the expression of novel targets verses expression in conventional models.

An adaptable, fit-for-purpose screening approach with high-throughput capability to determine speed of action and stage specificity of anti-malarial compounds.

A revamped in vitro compound identification and activity profiling approach is required to meet the large unmet need for new anti-malarial drugs to combat parasite drug resistance. Although compound hit identification utilizing high-throughput screening of large compound libraries is well established, the ability to rapidly prioritize such large numbers for further development is limited. Determining the speed of action of anti-malarial drug candidates is a vital component of malaria drug discovery, which currently occurs predominantly in lead optimization and development. This is due in part to the capacity of current methods which have low throughput due to the complexity and labor intensity of the approaches. Here, we provide an adaptable screening paradigm utilizing automated high content imaging, including the development of an automated schizont maturation assay, which collectively can identify anti-malarial compounds, classify activity into fast and slow acting, and provide an indication of the parasite stage specificity, with high-throughput capability. By frontloading these critical biological parameters much earlier in the drug discovery pipeline, it has the potential to reduce lead compound attrition rates later in the development process. The capability of the approach in its alternative formats is demonstrated using three Medicines for Malaria Venture open access compound "boxes," namely Pathogen Box (malaria set-125 compounds), Global Health Priority Box [Malaria Box 2 (80 compounds) and zoonotic neglected diseases (80 compounds)], and the Pandemic Response Box (400 compounds). From a total of 685 compounds tested, 79 were identified as having fast ring-stage-specific activity comparable to that of artemisinin and therefore of high priority for further consideration and development.

Bioefficacy, chromatographic profiling and drug-likeness analysis of flavonoids and terpenoids as potential inhibitors of H1N1 influenza viral proteins

Considering medicinal plants, natural products present in these plants are the best sources of medications for combating viral infection. The possible drug target against viral H1N1 influenza proteins lead to identification of selected secondary metabolites from potential plants Tinospora cordifolia, Ocimum sanctum, and Piper nigrum. On analysis of in vitro cell based antiviral activity of the selected plant extracts, an indication for a possible lead compound against neuraminidase activity was evident. Potent ligands were selected using drug docking and ADMET analysis, and the screened lead metabolites were ultimately identified as terpenoid (Columbin) and, flavonoid (Cubebin, and Apigenin). Among the selected ligands, the drug binding activity of Cubebin with all the 6 proteins of H1N1 influenza type A virus, HA (4r8w), NA (4qn7), M2 (3lbw), PA (4wsb), PB1 (2znl) and PB2 (3wil), was pronounced. In addition, physicochemical and pharmacokinetic parameters linked to absorption, distribution, metabolism, excretion and toxicity (ADMET) have been evaluated and corroborate with our in vitro results. Molecular dynamics modelling indicated Cubebin can be a potential phytochemical in a drug discovery pipeline for the development of neuraminidase inhibitors. Further studies can provide a possibility for an alternative therapy against Influenza viruses.

AI-Driven Drug Discovery: Accelerating the Development of Novel Therapeutics in Biopharmaceuticals

Artificial Intelligence (AI) has emerged as a transformative force in drug discovery, revolutionizing the biopharmaceutical industry's approach to developing novel therapeutics. This paper provides a comprehensive overview of AI-driven drug discovery, focusing on its applications in accelerating the development of innovative treatments. We examine the fundamental AI technologies employed in drug discovery, including machine learning algorithms, deep learning architectures, and natural language processing techniques. The paper analyzes the integration of AI across various stages of the drug discovery pipeline, from target identification to clinical trial design, highlighting significant improvements in efficiency and accuracy. We explore the impact of big data on AI-driven drug discovery, discussing the challenges and opportunities presented by multi-omics data integration, electronic health records mining, and the need for data standardization. The study also addresses ethical considerations and regulatory challenges associated with AI implementation in drug development. Finally, we present emerging trends and prospects for AI in biopharmaceuticals, emphasizing the importance of collaborative ecosystems and the potential for AI to revolutionize personalized medicine. This review synthesizes current research and industry practices, providing insights into the transformative potential of AI in drug discovery and the challenges that lie ahead in realizing its full potential.

Targeting CCRL2 enhances therapeutic outcomes in a tuberculosis mouse model.

Tuberculosis (TB) remains among the leading infectious causes of death. Due to the limited number of antimicrobials in the TB drug discovery pipeline, interest has developed in host-directed approaches to improve TB treatment outcomes. C-C motif chemokine-like receptor 2 (CCRL2) is a unique seven-transmembrane domain receptor that is upregulated by inflammatory signals and mediates leucocyte migration. However, little is known about its role in the setting of TB infection. Here, we show that Mycobacterium tuberculosis (Mtb) infection increases CCRL2 protein expression in macrophages and in mouse lungs. To target selectively CCRL2-expressing cells in vivo, we developed a novel mouse anti-CCRL2 antibody-drug conjugate (ADC) linked with the cytotoxic drug SG3249. We tested its adjunctive therapeutic efficacy against TB when combined with the first-line regimen for drug-susceptible TB (isoniazid, rifampin, pyrazinamide, ethambutol; RHZE). The anti-CCRL2 ADC treatment potentiated RHZE efficacy in Mtb-infected mice and decreased gross lung inflammation. CCRL2 expression in lung dendritic cells and alveolar macrophages was lower in mice receiving anti-CCRL2 ADC treatment + RHZE compared to those receiving RHZE alone or the control group, although the total innate cell populations did not differ across treatment groups. Interestingly, neutrophils were completely absent in the anti-CCRL2 ADC treatment + RHZE group, unlike in the other treatment groups. IFN-γ+ and IL17-Α+ T-cell responses, which are associated with optimal TB control, were also elevated in the anti-CCRL2 ADC treatment + RHZE group. Collectively, our findings suggest that CCRL2-targeting approaches may improve TB treatment outcomes, possibly through selective killing of Mtb-infected innate immune cells.

Computationally accelerated identification of P-glycoprotein inhibitors.

Overexpression of the polyspecific efflux transporter, P-glycoprotein (P-gp, MDR1, ABCB1 ), is a major mechanism by which cancer cells acquire multidrug resistance (MDR), the resistance to diverse chemotherapeutic drugs. Inhibiting drug transport by P-gp can resensitize cancer cells to chemotherapy, but there are no P-gp inhibitors available to patients. Clinically unsuccessful P-gp inhibitors tend to bind at the pump's transmembrane drug binding domains and are often P-gp transport substrates, resulting in lowered intracellular concentration of the drug and altered pharmacokinetics. In prior work, we used computationally accelerated drug discovery to identify novel P-gp inhibitors that target the pump's cytoplasmic nucleotide binding domains. Our first-draft study provided conclusive evidence that the nucleotide binding domains of P-gp are viable targets for drug discovery. Here we develop an enhanced, computationally accelerated drug discovery pipeline that expands upon our prior work by iteratively screening compounds against multiple conformations of P-gp with molecular docking. Targeted molecular dynamics simulations with our homology model of human P-gp were used to generate docking receptors in conformations mimicking a putative drug transport cycle. We offset the increased computational complexity using custom Tanimoto chemical datasets, which maximize the chemical diversity of ligands screened by docking. Using our expanded, virtual-assisted pipeline, we identified nine novel P-gp inhibitors that reverse MDR in two types of P-gp overexpressing human cancer cell lines, reflecting a 13.4% hit rate. Of these inhibitors, all were non-toxic to non-cancerous human cells, and six were not likely to be transport substrates of P-gp. Our novel P-gp inhibitors are chemically diverse and are good candidates for lead optimization. Our results demonstrate that the nucleotide binding domains of P-gp are an underappreciated target in the effort to reverse P-gp-mediated multidrug resistance in cancer.

Deciphering Host-Parasite Interplay in Leishmania Infection through a One Health View of Proteomics Studies on Drug Resistance.

Recent efforts in the study of vector-borne parasitic diseases (VBPDs) have emphasized an increased consideration for preventing drug resistance and promoting the environmental safety of drugs, from the beginning of the drug discovery pipeline. The intensive use of the few available antileishmanial drugs has led to the spreading of hyper-resistant Leishmania infantum strains, resulting in a chronic burden of the disease. In the present work, we have investigated the biochemical mechanisms of resistance to antimonials, paromomycin, and miltefosine in three drug-resistant parasitic strains from human clinical isolates, using a whole-cell mass spectrometry proteomics approach. We identified 14 differentially expressed proteins that were validated with their transcripts. Next, we employed functional association networks to identify parasite-specific proteins as potential targets for novel drug discovery studies. We used SeqAPASS analysis to predict susceptibility based on the evolutionary conservation of protein drug targets across species. MATH-domain-containing protein, adenosine triphosphate (ATP)-binding cassette B2, histone H4, calpain-like cysteine peptidase, and trypanothione reductase emerged as top candidates. Overall, this work identifies new biological targets for designing drugs to prevent the development of Leishmania drug resistance, while aligning with One Health principles that emphasize the interconnected health of people, animals, and ecosystems.

Fluorescence-Based HTS Assays for Ion Channel Modulation in Drug Discovery Pipelines.

Ion channels represent a druggable family of transmembrane pore-forming proteins with important (patho)physiological functions. While electrophysiological measurement (manual patch clamp) remains the only direct method for detection of ion currents, it is a labor-intensive technique. Although automated patch clamp instruments have become available to date, their high costs limit their use to large pharma companies or commercial screening facilities. Therefore, fluorescence-based assays are particularly important for initial screening of compound libraries. Despite their numerous disadvantages, they are highly amenable to high-throughput screening and in many cases, no sophisticated instrumentation or materials are required. These features predispose them for implementation in early phases of drug discovery pipelines (hit identification), even in an academic environment. This review summarizes the advantages and pitfalls of individual methodological approaches for identification of ion channel modulators employing fluorescent probes (i.e., membrane potential and ion flux assays) with emphasis on practical aspects of their adaptation to high-throughput format.

ResGAT: Residual Graph Attention Networks for molecular property prediction

Molecular property prediction is an important step in the drug discovery pipeline. Numerous computational methods have been developed to predict a wide range of molecular properties. While recent approaches have shown promising results, no single architecture can comprehensively address all tasks, making this area persistently challenging and requiring substantial time and effort. Beyond traditional machine learning and deep learning architectures for regular data, several deep learning architectures have been designed for graph-structured data to overcome the limitations of conventional methods. Utilizing graph-structured data in quantitative structure–activity relationship (QSAR) modeling allows models to effectively extract unique features, especially where connectivity information is crucial. In our study, we developed residual graph attention networks (ResGAT), a deep learning architecture for molecular graph-structured data. This architecture is a combination of graph attention networks and shortcut connections to address both regression and classification problems. It is also customizable to adapt to various dataset sizes, enhancing the learning process based on molecular patterns. When tested multiple times with both random and scaffold sampling strategies on nine benchmark molecular datasets, QSAR models developed using ResGAT demonstrated stability and competitive performance compared to state-of-the-art methods.

A Drug Discovery Pipeline for MAPK/ERK Pathway Inhibitors in Caenorhabditis elegans.

Many tumors depend on MAPK/ERK signaling to sustain growth, avoid cell death, and metastasize. We show that specific and clinically relevant MAPK/ERK signaling inhibitors can be discovered in vivo with a high-throughput screening pipeline in small animals.

Accelerating drug discovery and repurposing by combining transcriptional signature connectivity with docking.

We present an in silico approach for drug discovery, dubbed connectivity enhanced structure activity relationship (ceSAR). Building on the landmark LINCS library of transcriptional signatures of drug-like molecules and gene knockdowns, ceSAR combines cheminformatic techniques with signature concordance analysis to connect small molecules and their targets and further assess their biophysical compatibility using molecular docking. Candidate compounds are first ranked in a target structure-independent manner, using chemical similarity to LINCS analogs that exhibit transcriptomic concordance with a target gene knockdown. Top candidates are subsequently rescored using docking simulations and machine learning-based consensus of the two approaches. Using extensive benchmarking, we show that ceSAR greatly reduces false-positive rates, while cutting run times by multiple orders of magnitude and further democratizing drug discovery pipelines. We further demonstrate the utility of ceSAR by identifying and experimentally validating inhibitors of BCL2A1, an important antiapoptotic target in melanoma and preterm birth-associated inflammation.

BuildAMol: a versatile Python toolkit for fragment-based molecular design

In recent years computational methods for molecular modeling have become a prime focus of computational biology and cheminformatics. Many dedicated systems exist for modeling specific classes of molecules such as proteins or small drug-like ligands. These are often heavily tailored toward the automated generation of molecular structures based on some meta-input by the user and are not intended for expert-driven structure assembly. Dedicated manual or semi-automated assembly software tools exist for a variety of molecule classes but are limited in the scope of structures they can produce. In this work we present BuildAMol, a highly flexible and extendable, general-purpose fragment-based molecular assembly toolkit. Written in Python and featuring a well-documented, user-friendly API, BuildAMol empowers researchers with a framework for detailed manual or semi-automated construction of diverse molecular models. Unlike specialized software, BuildAMol caters to a broad range of applications. We demonstrate its versatility across various use cases, encompassing generating metal complexes or the modeling of dendrimers or integrated into a drug discovery pipeline. By providing a robust foundation for expert-driven model building, BuildAMol holds promise as a valuable tool for the continuous integration and advancement of powerful deep learning techniques.Scientific contributionBuildAMol introduces a cutting-edge framework for molecular modeling that seamlessly blends versatility with user-friendly accessibility. This innovative toolkit integrates modeling, modification, optimization, and visualization functions within a unified API, and facilitates collaboration with other cheminformatics libraries. BuildAMol, with its shallow learning curve, serves as a versatile tool for various molecular applications while also laying the groundwork for the development of specialized software tools, contributing to the progress of molecular research and innovation.

Abstract Or121: Druggable Genome CRISPRi/a Screens in iPSC-Derived Cardiomyocytes Identify Therapeutic Targets for Doxorubicin-Induced Cardiotoxicity

Introduction: Doxorubicin is a highly effective chemotherapy drug whose therapeutic utility is significantly hampered by its life-threatening adverse cardiac effects. The need for strategies to mitigate these cardiotoxic effects without diminishing its efficacy against cancer is critical in cardio-oncology. Hypothesis: We hypothesize that through an integrated drug discovery pipeline, utilizing patient iPSC-derived cardiomyocytes (iCMs) and CRISPRi/a bidirectional pooled screens combined with small molecule screening and molecular docking, we can identify therapeutic targets that ameliorate doxorubicin-induced cardiotoxicity (DIC) while preserving its anticancer efficacy. Methods: Our approach included patient iPSC-derived cardiomyocytes (iCMs), CRISPRi/a bidirectional pooled screens targeting druggable genome (~2000 genes) to identify candidate genes implicated in DIC, and small molecule screening alongside molecular docking to find potential inhibitors. Notably, carbonic anhydrase 12 (CA12) was identified as a primary hit for its role in DIC. Results: The CRISPRi/a screens revealed several novel genes, including CA12, as potential contributors to DIC. Genetic inhibition of CA12 in iCMs enhanced cell survival, preserved sarcomere structure, and improved contractile function and calcium handling, effectively protecting against DIC. Furthermore, we discovered that a CA12 antagonist significantly attenuates DIC in both iCMs and mouse models. Mechanistic studies suggested that doxorubicin triggers CA12 activation, leading to a metabolic change towards glycolysis in cardiomyocytes, thereby exacerbating DIC by disrupting cellular metabolism and cardiac functions. Conclusions: Our findings propose a novel therapeutic avenue to safeguard against doxorubicin-induced cardiotoxicity through the targeting of CA12, offering new direction for the development of more refined anticancer therapies that minimize cardiotoxic risk. This study sets the stage of combining CRISPRi/a screen with iPSC technology for future drug discovery endeavors aimed at eliminating off-target cardiotoxicities of chemotherapy agents.

Meta Learning With Graph Attention Networks for Low-Data Drug Discovery.

Finding candidate molecules with favorable pharmacological activity, low toxicity, and proper pharmacokinetic properties is an important task in drug discovery. Deep neural networks have made impressive progress in accelerating and improving drug discovery. However, these techniques rely on a large amount of label data to form accurate predictions of molecular properties. At each stage of the drug discovery pipeline, usually, only a few biological data of candidate molecules and derivatives are available, indicating that the application of deep neural networks for low-data drug discovery is still a formidable challenge. Here, we propose a meta learning architecture with graph attention network, Meta-GAT, to predict molecular properties in low-data drug discovery. The GAT captures the local effects of atomic groups at the atom level through the triple attentional mechanism and implicitly captures the interactions between different atomic groups at the molecular level. GAT is used to perceive molecular chemical environment and connectivity, thereby effectively reducing sample complexity. Meta-GAT further develops a meta learning strategy based on bilevel optimization, which transfers meta knowledge from other attribute prediction tasks to low-data target tasks. In summary, our work demonstrates how meta learning can reduce the amount of data required to make meaningful predictions of molecules in low-data scenarios. Meta learning is likely to become the new learning paradigm in low-data drug discovery. The source code is publicly available at: https://github.com/lol88/Meta-GAT.

A hybrid quantum computing pipeline for real world drug discovery

Quantum computing, with its superior computational capabilities compared to classical approaches, holds the potential to revolutionize numerous scientific domains, including pharmaceuticals. However, the application of quantum computing for drug discovery has primarily been limited to proof-of-concept studies, which often fail to capture the intricacies of real-world drug development challenges. In this study, we diverge from conventional investigations by developing a hybrid quantum computing pipeline tailored to address genuine drug design problems. Our approach underscores the application of quantum computation in drug discovery and propels it towards more scalable system. We specifically construct our versatile quantum computing pipeline to address two critical tasks in drug discovery: the precise determination of Gibbs free energy profiles for prodrug activation involving covalent bond cleavage, and the accurate simulation of covalent bond interactions. This work serves as a pioneering effort in benchmarking quantum computing against veritable scenarios encountered in drug design, especially the covalent bonding issue present in both of the case studies, thereby transitioning from theoretical models to tangible applications. Our results demonstrate the potential of a quantum computing pipeline for integration into real world drug design workflows.

Machine learning in preclinical drug discovery.

Drug-discovery and drug-development endeavors are laborious, costly and time consuming. These programs can take upward of 12 years and cost US $2.5 billion, with a failure rate of more than 90%. Machine learning (ML) presents an opportunity to improve the drug-discovery process. Indeed, with the growing abundance of public and private large-scale biological and chemical datasets, ML techniques are becoming well positioned as useful tools that can augment the traditional drug-development process. In this Perspective, we discuss the integration of algorithmic methods throughout the preclinical phases of drug discovery. Specifically, we highlight an array of ML-based efforts, across diverse disease areas, to accelerate initial hit discovery, mechanism-of-action (MOA) elucidation and chemical property optimization. With advances in the application of ML across diverse therapeutic areas, we posit that fully ML-integrated drug-discovery pipelines will define the future of drug-development programs.

Parallelized Droplet Vitrification Enables Single-Run Vitrification of the Whole Rat Liver Hepatocyte Yield.

Drug discovery pipelines rely on the availability of isolated primary hepatocytes for investigating potential hepatotoxicity prior to clinical application. These hepatocytes are typically isolated from livers rejected for transplantation and subsequently cryopreserved for later usage. The gold-standard cryopreservation technique, slow-freezing, is a labor-intensive process, with significant post-storage viability loss. In this work, we introduce parallelized droplet vitrification, a technique for high-volumetric, rapid vitrification of suspended cells. We show the utility of this technique through the single-run vitrification of the whole-rate liver hepatocyte yield, resulting in a 1600% increase in single-batch vitrification and a 500% increase in droplet generation rate compared to previous droplet vitrification approaches. Additionally, we showed that these implementations maintained improved post-preservation outcomes in primary rat hepatocytes.