Biomarkers of aging for the identification and evaluation of longevity interventions.

With the rapid expansion of aging biology research, the identification and evaluation of longevity interventions in humans have become key goals of this field. Biomarkers of aging are critically important tools in achieving these objectives over realistic time frames. However, the current lack of standards and consensus on the properties of a reliable aging biomarker hinders their further development and validation for clinical applications. Here, we advance a framework for the terminology and characterization of biomarkers of aging, including classification and potential clinical use cases. We discuss validation steps and highlight ongoing challenges as potential areas in need of future research. This framework sets the stage for the development of valid biomarkers of aging and their ultimate utilization in clinical trials and practice.

Multiomic prediction of therapeutic targets for human diseases associated with protein phase separation.

The phenomenon of protein phase separation (PPS) underlies a wide range of cellular functions. Correspondingly, the dysregulation of the PPS process has been associated with numerous human diseases. To enable therapeutic interventions based on the regulation of this association, possible targets should be identified. For this purpose, we present an approach that combines the multiomic PandaOmics platform with the FuzDrop method to identify PPS-prone disease-associated proteins. Using this approach, we prioritize candidates with high PandaOmics and FuzDrop scores using a profiling method that accounts for a wide range of parameters relevant for disease mechanism and pharmacological intervention. We validate the differential phase separation behaviors of three predicted Alzheimer's disease targets (MARCKS, CAMKK2, and p62) in two cell models of this disease. Overall, the approach that we present generates a list of possible therapeutic targets for human diseases associated with the dysregulation of the PPS process.

PandaOmics: An AI-Driven Platform for Therapeutic Target and Biomarker Discovery.

PandaOmics is a cloud-based software platform that applies artificial intelligence and bioinformatics techniques to multimodal omics and biomedical text data for therapeutic target and biomarker discovery. PandaOmics generates novel and repurposed therapeutic target and biomarker hypotheses with the desired properties and is available through licensing or collaboration. Targets and biomarkers generated by the platform were previously validated in both in vitro and in vivo studies. PandaOmics is a core component of Insilico Medicine's Pharma.ai drug discovery suite, which also includes Chemistry42 for the de novo generation of novel small molecules, and inClinico─a data-driven multimodal platform that forecasts a clinical trial's probability of successful transition from phase 2 to phase 3. In this paper, we demonstrate how the PandaOmics platform can efficiently identify novel molecular targets and biomarkers for various diseases.

Discovery of 3-hydroxymethyl-azetidine derivatives as potent polymerase theta inhibitors.

Inhibition of the low fidelity DNA polymerase Theta (Polθ) is emerging as an attractive, synthetic-lethal antitumor strategy in BRCA-deficient tumors. Here we report the AI-enabled development of 3-hydroxymethyl-azetidine derivatives as a novel class of Polθ inhibitors featuring central scaffolding rings. Structure-based drug design first identified A7 as a lead compound, which was further optimized to the more potent derivative B3 and the metabolically stable deuterated compound C1. C1 exhibited significant antiproliferative properties in DNA repair-compromised cells and demonstrated favorable pharmacokinetics, showcasing that 3-hydroxymethyl-azetidine is an effective bio-isostere of pyrrolidin-3-ol and emphasizing the potential of AI in medicinal chemistry for precise molecular modifications.

Discovery of novel MAT2A inhibitors by an allosteric site-compatible fragment growing approach.

The methionine adenosyltransferase MAT2A catalyzes the synthesis ofthe methyl donor S-adenosylmethionine (SAM) and thereby regulates critical aspects of metabolism and transcription. Aberrant MAT2A function can lead to metabolic and transcriptional reprogramming of cancer cells, and MAT2A has been shown to promote survival of MTAP-deficient tumors, a genetic alteration that occurs in âˆ¼ 13 % of all tumors. Thus, MAT2A holds great promise as a novel anticancer target. Here, we report a novel series of MAT2A inhibitors generated by a fragment growing approach from AZ-28, a low-molecular weight MAT2A inhibitor with promising pre-clinical properties. X-ray co-crystal structure revealed that compound 7 fully occupies the allosteric pocket of MAT2A as a single molecule mimicking MAT2B. By introducing additional backbone interactions and rigidifying the requisite linker extensions, we generated compound 8, which exhibited single digit nanomolar enzymatic and sub-micromolar cellular inhibitory potency for MAT2A.

Discovery of macrocyclic CDK2/4/6 inhibitors with improved potency and DMPK properties through a highly efficient macrocyclic drug design platform.

Cyclin-dependent kinases (CDKs) are critical cell cycle regulators that are often overexpressed in tumors, making them promising targets for anti-cancer therapies. Despite substantial advancements in optimizing the selectivity and drug-like properties of CDK inhibitors, safety of multi-target inhibitors remains a significant challenge. Macrocyclization is a promising drug discovery strategy to improve the pharmacological properties of existing compounds. Here we report the development of a macrocyclization platform that enabled the highly efficient discovery of a novel, macrocyclic CDK2/4/6 inhibitor from an acyclic precursor (NUV422). Using dihedral angle scan and structure-based, computer-aided drug design to select an optimal ring-closing site and linker length for the macrocycle, we identified compound 8 as a potent new CDK2/4/6 inhibitor with optimized cellular potency and safety profile compared to NUV422. Our platform leverages both experimentally-solved as well as generative chemistry-derived macrocyclic structures and can be deployed to streamline the design of macrocyclic new drugs from acyclic starting compounds, yielding macrocyclic compounds with enhanced potency and improved drug-like properties.

Skinly: A novel handheld IoT device for validating biophysical skin characteristics.

BACKGROUND: Recent advancements in artificial intelligence have revolutionized dermatological diagnostics. These technologies, particularly machine learning (ML), including deep learning (DL), have shown accuracy equivalent or even superior to human experts in diagnosing skin conditions like melanoma. With the integration of ML, including DL, the development of at home skin analysis devices has become feasible. To this end, we introduced the Skinly system, a handheld device capable of evaluating various personal skin characteristics noninvasively. MATERIALS AND METHODS: Equipped with a moisture sensor and a multi-light-source camera, Skinly can assess age-related skin parameters and specific skin properties. Utilizing state-of-the-art DL, Skinly processed vast amounts of images efficiently. The Skinly system's efficacy was validated both in the lab and at home, comparing its results to established "gold standard" methods. RESULTS: Our findings revealed that the Skinly device can accurately measure age-associated parameters, that is, facial age, skin evenness, and wrinkles. Furthermore, Skinly produced data consistent with established devices for parameters like glossiness, skin tone, redness, and porphyrin levels. A separate study was conducted to evaluate the effects of two moisturizing formulations on skin hydration in laboratory studies with standard instrumentation and at home with Skinly. CONCLUSION: Thanks to its capability for multi-parameter measurements, the Skinly device, combined with its smartphone application, holds the potential to replace more expensive, time-consuming diagnostic tools. Collectively, the Skinly device opens new avenues in dermatological research, offering a reliable, versatile tool for comprehensive skin analysis.

Chemistry42: An AI-Driven Platform for Molecular Design and Optimization.

Chemistry42 is a software platform for de novo small molecule design and optimization that integrates Artificial Intelligence (AI) techniques with computational and medicinal chemistry methodologies. Chemistry42 efficiently generates novel molecular structures with optimized properties validated in both in vitro and in vivo studies and is available through licensing or collaboration. Chemistry42 is the core component of Insilico Medicine's Pharma.ai drug discovery suite. Pharma.ai also includes PandaOmics for target discovery and multiomics data analysis, and inClinico─a data-driven multimodal forecast of a clinical trial's probability of success (PoS). In this paper, we demonstrate how the platform can be used to efficiently find novel molecular structures against DDR1 and CDK20.

Exploring the Advantages of Quantum Generative Adversarial Networks in Generative Chemistry.

De novo drug design with desired biological activities is crucial for developing novel therapeutics for patients. The drug development process is time- and resource-consuming, and it has a low probability of success. Recent advances in machine learning and deep learning technology have reduced the time and cost of the discovery process and therefore, improved pharmaceutical research and development. In this paper, we explore the combination of two rapidly developing fields with lead candidate discovery in the drug development process. First, artificial intelligence has already been demonstrated to successfully accelerate conventional drug design approaches. Second, quantum computing has demonstrated promising potential in different applications, such as quantum chemistry, combinatorial optimizations, and machine learning. This article explores hybrid quantum-classical generative adversarial networks (GAN) for small molecule discovery. We substituted each element of GAN with a variational quantum circuit (VQC) and demonstrated the quantum advantages in the small drug discovery. Utilizing a VQC in the noise generator of a GAN to generate small molecules achieves better physicochemical properties and performance in the goal-directed benchmark than the classical counterpart. Moreover, we demonstrate the potential of a VQC with only tens of learnable parameters in the generator of GAN to generate small molecules. We also demonstrate the quantum advantage of a VQC in the discriminator of GAN. In this hybrid model, the number of learnable parameters is significantly less than the classical ones, and it can still generate valid molecules. The hybrid model with only tens of training parameters in the quantum discriminator outperforms the MLP-based one in terms of both generated molecule properties and the achieved KL divergence. However, the hybrid quantum-classical GANs still face challenges in generating unique and valid molecules compared to their classical counterparts.

Discovery of novel and selective SIK2 inhibitors by the application of AlphaFold structures and generative models.

Salt-inducible kinase 2 (SIK2) has been recognized as a potential target for anti-inflammation and anti-cancer therapy. In this paper, based on the binding pose of the reported compound (GLPG-3970, 3) with AlphaFold protein structure, a series of hinge cores were generated via AI-generative models (Chemistry42). After the molecular docking, synthesis, and biological evaluation, a hit molecule (7f) targeting SIK2 was obtained with a novel scaffold. Further SAR exploration led to the discovery of compound 8g with superior potency against SIK2 compared with the reported inhibitors. Furthermore, 8g also demonstrated excellent selectivity over other AMPK kinases, favorable in vitro ADMET profiles and decent cellular activities. This work provides an alternative approach to the discovery of novel and selective kinase inhibitors.

Application of liquid biopsy as multi-functional biomarkers in head and neck cancer.

Head and neck squamous cell carcinoma (HNSCC) is a molecularly heterogeneous disease, with a 5-year survival rate that still hovers at ~60% despite recent advancements. The advanced stage upon diagnosis, limited success with effective targeted therapy and lack of reliable biomarkers are among the key factors underlying the marginally improved survival rates over the decades. Prevention, early detection and biomarker-driven treatment adaptation are crucial for timely interventions and improved clinical outcomes. Liquid biopsy, analysis of tumour-specific biomarkers circulating in bodily fluids, is a rapidly evolving field that may play a striking role in optimising patient care. In recent years, significant progress has been made towards advancing liquid biopsies for non-invasive early cancer detection, prognosis, treatment adaptation, monitoring of residual disease and surveillance of recurrence. While these emerging technologies have immense potential to improve patient survival, numerous methodological and biological limitations must be overcome before their implementation into clinical practice. This review outlines the current state of knowledge on various types of liquid biopsies in HNSCC, and their potential applications for diagnosis, prognosis, grading treatment response and post-treatment surveillance. It also discusses challenges associated with the clinical applicability of liquid biopsies and prospects of the optimised approaches in the management of HNSCC.

COVIDomic: A multi-modal cloud-based platform for identification of risk factors associated with COVID-19 severity.

Coronavirus disease 2019 (COVID-19) is an acute infection of the respiratory tract that emerged in December 2019 in Wuhan, China. It was quickly established that both the symptoms and the disease severity may vary from one case to another and several strains of SARS-CoV-2 have been identified. To gain a better understanding of the wide variety of SARS-CoV-2 strains and their associated symptoms, thousands of SARS-CoV-2 genomes have been sequenced in dozens of countries. In this article, we introduce COVIDomic, a multi-omics online platform designed to facilitate the analysis and interpretation of the large amount of health data collected from patients with COVID-19. The COVIDomic platform provides a comprehensive set of bioinformatic tools for the multi-modal metatranscriptomic data analysis of COVID-19 patients to determine the origin of the coronavirus strain and the expected severity of the disease. An integrative analytical workflow, which includes microbial pathogens community analysis, COVID-19 genetic epidemiology and patient stratification, allows to analyze the presence of the most common microbial organisms, their antibiotic resistance, the severity of the infection and the set of the most probable geographical locations from which the studied strain could have originated. The online platform integrates a user friendly interface which allows easy visualization of the results. We envision this tool will not only have immediate implications for management of the ongoing COVID-19 pandemic, but will also improve our readiness to respond to other infectious outbreaks.

Fetal mitochondrial DNA in maternal plasma in surrogate pregnancies: Detection and topology.

OBJECTIVES: Due to the maternally-inherited nature of mitochondrial DNA (mtDNA), there is a lack of information regarding fetal mtDNA in the plasma of pregnant women. We aim to explore the presence and topologic forms of circulating fetal and maternal mtDNA molecules in surrogate pregnancies. METHODS: Genotypic differences between fetal and surrogate maternal mtDNA were used to identify the fetal and maternal mtDNA molecules in plasma. Plasma samples were obtained from the surrogate pregnant mothers. Using cleavage-end signatures of BfaI restriction enzyme, linear and circular mtDNA molecules in maternal plasma could be differentiated. RESULTS: Fetal-derived mtDNA molecules were mainly linear (median: 88%; range: 80%-96%), whereas approximately half of the maternal-derived mtDNA molecules were circular (median: 51%; range: 42%-60%). The fetal DNA fraction of linear mtDNA was lower (median absolute difference: 9.8%; range: 1.1%-27%) than that of nuclear DNA (median: 20%; range: 9.7%-35%). The fetal-derived linear mtDNA molecules were shorter than the maternal-derived ones. CONCLUSION: Fetal mtDNA is present in maternal plasma, and consists mainly of linear molecules. Surrogate pregnancies represent a valuable clinical scenario for exploring the biology and potential clinical applications of circulating mtDNA, for example, for pregnancies conceived following mitochondrial replacement therapy.

Medicinal Chemists versus Machines Challenge: What Will It Take to Adopt and Advance Artificial Intelligence for Drug Discovery?

The field of artificial intelligence (AI) for generative chemistry is reaching the maturity stage, shifting the focus from the novelty of the algorithms to the quality of the generated molecules. To ensure continued evolution of AI technologies, we propose a series of challenges of increasing complexity by comparing and combining the machine and human intelligence in medicinal chemistry.

Adversarial Threshold Neural Computer for Molecular de Novo Design.

In this article, we propose the deep neural network Adversarial Threshold Neural Computer (ATNC). The ATNC model is intended for the de novo design of novel small-molecule organic structures. The model is based on generative adversarial network architecture and reinforcement learning. ATNC uses a Differentiable Neural Computer as a generator and has a new specific block, called adversarial threshold (AT). AT acts as a filter between the agent (generator) and the environment (discriminator + objective reward functions). Furthermore, to generate more diverse molecules we introduce a new objective reward function named Internal Diversity Clustering (IDC). In this work, ATNC is tested and compared with the ORGANIC model. Both models were trained on the SMILES string representation of the molecules, using four objective functions (internal similarity, Muegge druglikeness filter, presence or absence of sp3-rich fragments, and IDC). The SMILES representations of 15K druglike molecules from the ChemDiv collection were used as a training data set. For the different functions, ATNC outperforms ORGANIC. Combined with the IDC, ATNC generates 72% of valid and 77% of unique SMILES strings, while ORGANIC generates only 7% of valid and 86% of unique SMILES strings. For each set of molecules generated by ATNC and ORGANIC, we analyzed distributions of four molecular descriptors (number of atoms, molecular weight, logP, and tpsa) and calculated five chemical statistical features (internal diversity, number of unique heterocycles, number of clusters, number of singletons, and number of compounds that have not been passed through medicinal chemistry filters). Analysis of key molecular descriptors and chemical statistical features demonstrated that the molecules generated by ATNC elicited better druglikeness properties. We also performed in vitro validation of the molecules generated by ATNC; results indicated that ATNC is an effective method for producing hit compounds.

3D Molecular Representations Based on the Wave Transform for Convolutional Neural Networks.

Convolutional neural networks (CNN) have been successfully used to handle three-dimensional data and are a natural match for data with spatial structure such as 3D molecular structures. However, a direct 3D representation of a molecule with atoms localized at voxels is too sparse, which leads to poor performance of the CNNs. In this work, we present a novel approach where atoms are extended to fill other nearby voxels with a transformation based on the wave transform. Experimenting on 4.5 million molecules from the Zinc database, we show that our proposed representation leads to better performance of CNN-based autoencoders than either the voxel-based representation or the previously used Gaussian blur of atoms and then successfully apply the new representation to classification tasks such as MACCS fingerprint prediction.

Entangled Conditional Adversarial Autoencoder for de Novo Drug Discovery.

Modern computational approaches and machine learning techniques accelerate the invention of new drugs. Generative models can discover novel molecular structures within hours, while conventional drug discovery pipelines require months of work. In this article, we propose a new generative architecture, entangled conditional adversarial autoencoder, that generates molecular structures based on various properties, such as activity against a specific protein, solubility, or ease of synthesis. We apply the proposed model to generate a novel inhibitor of Janus kinase 3, implicated in rheumatoid arthritis, psoriasis, and vitiligo. The discovered molecule was tested in vitro and showed good activity and selectivity.

Reinforced Adversarial Neural Computer for de Novo Molecular Design.

In silico modeling is a crucial milestone in modern drug design and development. Although computer-aided approaches in this field are well-studied, the application of deep learning methods in this research area is at the beginning. In this work, we present an original deep neural network (DNN) architecture named RANC (Reinforced Adversarial Neural Computer) for the de novo design of novel small-molecule organic structures based on the generative adversarial network (GAN) paradigm and reinforcement learning (RL). As a generator RANC uses a differentiable neural computer (DNC), a category of neural networks, with increased generation capabilities due to the addition of an explicit memory bank, which can mitigate common problems found in adversarial settings. The comparative results have shown that RANC trained on the SMILES string representation of the molecules outperforms its first DNN-based counterpart ORGANIC by several metrics relevant to drug discovery: the number of unique structures, passing medicinal chemistry filters (MCFs), Muegge criteria, and high QED scores. RANC is able to generate structures that match the distributions of the key chemical features/descriptors (e.g., MW, logP, TPSA) and lengths of the SMILES strings in the training data set. Therefore, RANC can be reasonably regarded as a promising starting point to develop novel molecules with activity against different biological targets or pathways. In addition, this approach allows scientists to save time and covers a broad chemical space populated with novel and diverse compounds.

Aging Chart: a community resource for rapid exploratory pathway analysis of age-related processes.

Aging research is a multi-disciplinary field encompassing knowledge from many areas of basic, applied and clinical research. Age-related processes occur on molecular, cellular, tissue, organ, system, organismal and even psychological levels, trigger the onset of multiple debilitating diseases and lead to a loss of function, and there is a need for a unified knowledge repository designed to track, analyze and visualize the cause and effect relationships and interactions between the many elements and processes on all levels. Aging Chart (http://agingchart.org/) is a new, community-curated collection of aging pathways and knowledge that provides a platform for rapid exploratory analysis. Building on an initial content base constructed by a team of experts from peer-reviewed literature, users can integrate new data into biological pathway diagrams for a visible, intuitive, top-down framework of aging processes that fosters knowledge-building and collaboration. As the body of knowledge in aging research is rapidly increasing, an open visual encyclopedia of aging processes will be useful to both the new entrants and experts in the field.

druGAN: An Advanced Generative Adversarial Autoencoder Model for de Novo Generation of New Molecules with Desired Molecular Properties in Silico.

Deep generative adversarial networks (GANs) are the emerging technology in drug discovery and biomarker development. In our recent work, we demonstrated a proof-of-concept of implementing deep generative adversarial autoencoder (AAE) to identify new molecular fingerprints with predefined anticancer properties. Another popular generative model is the variational autoencoder (VAE), which is based on deep neural architectures. In this work, we developed an advanced AAE model for molecular feature extraction problems, and demonstrated its advantages compared to VAE in terms of (a) adjustability in generating molecular fingerprints; (b) capacity of processing very large molecular data sets; and (c) efficiency in unsupervised pretraining for regression model. Our results suggest that the proposed AAE model significantly enhances the capacity and efficiency of development of the new molecules with specific anticancer properties using the deep generative models.