Discovery Applications Research Articles

Multimodal Social Sensing for the Spatio-Temporal Evolution and Assessment of Nature Disasters.

Social sensing, using humans as sensors to collect disaster data, has emerged as a timely, cost-effective, and reliable data source. However, research has focused on the textual data. With advances in information technology, multimodal data such as images and videos are now shared on media platforms, aiding in-depth analysis of social sensing systems. This study proposed an analytical framework to extract disaster-related spatiotemporal information from multimodal social media data. Using a pre-trained multimodal neural network and a location entity recognition model, the framework integrates disaster semantics with spatiotemporal information, enhancing situational awareness. A case study of the April 2024 heavy rain event in Guangdong, China, using Weibo data, demonstrates that multimodal content correlates more strongly with rainfall patterns than textual data alone, offering a dynamic perception of disasters. These findings confirm the utility of multimodal social media data and offer a foundation for future research. The proposed framework offers valuable applications for emergency response, disaster relief, risk assessment, and witness discovery, and presents a viable approach for safety risk monitoring and early warning systems.

Rewireable Building Blocks for Enzyme-Powered DNA Computing Networks.

Neural networks enable the processing of large, complex data sets with applications in disease diagnosis, cell profiling, and drug discovery. Beyond electronic computers, neural networks have been implemented using programmable biomolecules such as DNA; this confers unique advantages, such as greater portability, electricity-free operation, and direct analysis of patterns of biomolecules in solution. Analogous to bottlenecks in electronic computers, the computing power of DNA-based neural networks is limited by the ability to add more computing units, i.e., neurons. This limitation exists because current architectures require many nucleic acids to model a single neuron. Each additional neuron compounds existing problems such as long assembly times, high background signal, and cross-talk between components. Here, we test three strategies to solve this limitation and improve the scalability of DNA-based neural networks: (i) enzymatic synthesis for high-purity neurons, (ii) spatial patterning of neuron clusters based on their network position, and (iii) encoding neuron connectivity on a universal single-stranded DNA backbone. We show that neurons implemented via these strategies activate quickly, with a high signal-to-background ratio and process-weighted inputs. We rewired our modular neurons to demonstrate basic neural network motifs such as cascading, fan-in, and fan-out circuits. Finally, we designed a prototype two-layer microfluidic device to automate the operation of our circuits. We envision that our proposed design will help scale DNA-based neural networks due to its modularity, simplicity of synthesis, and compatibility with various neural network architectures. This will enable portable computing power for applications in portable diagnostics, compact data storage, and autonomous decision making for lab-on-a-chips.

Recombinant Production of The Cyclotide Kalata B1 by Conditional Split Inteins.

This study describes the design, production, and characterization of a novel conditional intein system for the recombinant production of cyclic peptides. The system is based on two key features: (1) a promiscuous extein recognition site allowing cyclization of virtually any peptide, and (2) a secondary split site within the intein itself enabling triggered splicing at will. Two intein precursors were recombinantly expressed, purified, and then self-assembled in vitro to cyclize the model peptide kalata B1 (kB1). Cyclized kB1 was successfully purified, refolded, and characterized by mass spectrometry and NMR, demonstrating correct disulfide bond formation and identical structure to synthetic kB1. Importantly, the intein-derived kB1 retained full biological activity as evidenced by insect cell toxicity assays. This work establishes a versatile and efficient approach for intein-mediated protein cyclization with potential applications in bioengineering and peptide discovery.

TBHP Promotes Cross-Dehydrogenative Coupling of SF4 Alkynes with Tetrahydroisoquinolines under Copper Catalysis.

We present a viable approach for the cross-dehydrogenative coupling of Het-SF4-alkynes with tetrahydroisoquinolines under oxidative conditions, using TBHP and copper catalysts. These newly developed conditions boast enhanced yields and a more extensive range of substrates, demonstrating tolerance to various functional groups and addressing the limitations of earlier reports. Consequently, this method should increase opportunities for the exploration of SF4-containing compounds and their potential applications in drug discovery, materials science, and as alternatives to PFAS.

Single-molecule digital sizing of proteins in solution

The physical characterization of proteins in terms of their sizes, interactions, and assembly states is key to understanding their biological function and dysfunction. However, this has remained a difficult task because proteins are often highly polydisperse and present as multicomponent mixtures. Here, we address this challenge by introducing single-molecule microfluidic diffusional sizing (smMDS). This approach measures the hydrodynamic radius of single proteins and protein assemblies in microchannels using single-molecule fluorescence detection. smMDS allows for ultrasensitive sizing of proteins down to femtomolar concentrations and enables affinity profiling of protein interactions at the single-molecule level. We show that smMDS is effective in resolving the assembly states of protein oligomers and in characterizing the size of protein species within complex mixtures, including fibrillar protein aggregates and nanoscale condensate clusters. Overall, smMDS is a highly sensitive method for the analysis of proteins in solution, with wide-ranging applications in drug discovery, diagnostics, and nanobiotechnology.

Sample-Specific MS/MS Methods in High-Throughput Mass Spectrometry.

The drug discovery process increasingly relies on high-throughput sample analysis to accelerate the identification of viable drug candidates. Recently, chromatographic-free high-throughput mass spectrometry (HT-MS) technologies have emerged, significantly increasing sample readout speed and enabling the analysis of large sample sets. These HT-MS platforms continuously acquire data from various samples into a single data file, presenting challenges in applying distinctive data acquisition methods to specific samples. This study introduces a novel approach that integrates real-time sample loading status to activate sample-specific MS/MS data acquisition methods on the high-throughput acoustic ejection mass spectrometry platform. Effective method switching and high signal reproducibility were demonstrated across different data acquisition window durations in multiple reaction monitoring (MRM), high-resolution MRM (MRMHR), and information-dependent acquisition modes. This advancement provides a user-friendly and robust solution to the method-setting challenges of HT-MS, expanding the implementation of HT-MS platforms in drug discovery and other high-throughput analytical applications.

GTAM: a molecular pretraining model with geometric triangle awareness.

Molecular representation learning is pivotal for advancing deep learning applications in quantum chemistry and drug discovery. Existing methods for molecular representation learning often fall short of fully capturing the intricate interactions within chemical bonds of 2D topological graphs and the multifaceted effects of 3D geometric conformations. To overcome these challenges, we present a novel contrastive learning strategy for molecular representation learning, named Geometric Triangle Awareness Model (GTAM). This method integrates innovative molecular encoders for both 2D graphs and 3D conformations, enabling the accurate capture of geometric dependencies among edges in graph-based molecular structures. Furthermore, GTAM is bolstered by the development of two contrastive training objectives designed to facilitate the direct transfer of edge information between 2D topological graphs and 3D geometric conformations, enhancing the functionality of the molecular encoders. Through extensive evaluations on a range of 2D and 3D downstream tasks, our model has demonstrated superior performance over existing approaches. The test code and data of GTAM are available online at https://github.com/StellaHxy/GTAM.

Recent Advances in Nanocatalyzed One-Pot Sustainable Synthesis of Bioactive N, N-Heterocycles with Anticancer Activities: An Outlook of Medicinal Chemistry.

N-heterocycles represent a predominant and unique class of organic chemistry. They have received a lot of attention due to their important chemical, biomedical, and industrial uses. Food and Drug Administration (FDA) approved about 75% of drugs containing N-based heterocycles, which are currently available in the market. N-Heterocyclic compounds exist as the backbone of numerous natural products and act as crucial intermediates for the construction of pharmaceuticals, veterinary items, and agrochemicals frequently. Among N-based heterocyclic compounds, bioactive N,N-heterocycles constitute a broad spectrum of applications in modern drug discovery and development processes. Cefozopran (antibiotic), omeprazole (antiulcer), enviradine (antiviral), liarozole (anticancer), etc., are important drugs containing N,N-heterocycles. The synthesis of N,Nheterocyclic compounds under sustainable conditions is one of the most active fields because of their significant physiological and biological properties as well as synthetic utility. Current research is demanding the development of greener, cheaper, and milder protocols for the synthesis of N,N-heterocyclic compounds to save mother nature by avoiding toxic metal catalysts, extensive application of energy, and the excessive use of hazardous materials. Nanocatalysts play a profound role in sustainable synthesis because of their larger surface area, tiny size, and minimum energy; they are eco-friendly and safe, and they provide higher yields with selectivity in comparison to conventional catalysts. It is increasingly demanding research to design and synthesize novel bioactive compounds that may help to combat cancer since the major causes of death worldwide are due to cancer. Hence, the important uses of nanocatalysts for the one-pot synthesis of biologically potent N,N-heterocycles with anticancer activities have been presented in this review.

Advances in Deep Learning Assisted Drug Discovery Methods: A Self-review

: Artificial Intelligence is a field within computer science that endeavors to replicate the intricate structures and operational mechanisms inherent in the human brain. Machine learning is a subfield of artificial intelligence that focuses on developing models by analyzing training data. Deep learning is a distinct subfield within artificial intelligence, characterized by using models that depict geometric transformations across multiple layers. The deep learning has shown significant promise in various domains, including health and life sciences. In recent times, deep learning has demonstrated successful applications in drug discovery. In this self-review, we present recent methods developed with the aid of deep learning. The objective is to give a brief overview of the present cutting-edge advancements in drug discovery from our group. We have systematically discussed experimental evidence and proof of concept examples for the deep learning-based models developed, such as Deep- BindBC, DeepPep, and DeepBindRG. These developments not only shed light on the existing challenges but also emphasize the achievements and prospects for future drug discovery and development progress.

DTI-LM: language model powered drug-target interaction prediction.

The identification and understanding of drug-target interactions (DTIs) play a pivotal role in the drug discovery and development process. Sequence representations of drugs and proteins in computational model offer advantages such as their widespread availability, easier input quality control, and reduced computational resource requirements. These make them an efficient and accessible tools for various computational biology and drug discovery applications. Many sequence-based DTI prediction methods have been developed over the years. Despite the advancement in methodology, cold start DTI prediction involving unknown drug or protein remains a challenging task, particularly for sequence-based models. Introducing DTI-LM, a novel framework leveraging advanced pretrained language models, we harness their exceptional context-capturing abilities along with neighborhood information to predict DTIs. DTI-LM is specifically designed to rely solely on sequence representations for drugs and proteins, aiming to bridge the gap between warm start and cold start predictions. Large-scale experiments on four datasets show that DTI-LM can achieve state-of-the-art performance on DTI predictions. Notably, it excels in overcoming the common challenges faced by sequence-based models in cold start predictions for proteins, yielding impressive results. The incorporation of neighborhood information through a graph attention network further enhances prediction accuracy. Nevertheless, a disparity persists between cold start predictions for proteins and drugs. A detailed examination of DTI-LM reveals that language models exhibit contrasting capabilities in capturing similarities between drugs and proteins. Source code is available at: https://github.com/compbiolabucf/DTI-LM.

Multimodal Transformers and Their Applications in Drug Target Discovery for Aging and Age-Related Diseases.

Given the unprecedented rate of global aging, advancing aging research and drug discovery to support healthy and productive longevity is a pressing socioeconomic need. Holistic models of human and population aging that account for biomedical background, environmental context, and lifestyle choices are fundamental to address these needs, but integration of diverse data sources and large data sets into comprehensive models is challenging using traditional approaches. Recent advances in artificial intelligence and machine learning, and specifically multimodal transformer-based neural networks, have enabled the development of highly capable systems that can generalize across multiple data types. As such, multimodal transformers can generate systemic models of aging that can predict health status and disease risks, identify drivers, or breaks of physiological aging, and aid in target discovery against age-related disease. The unprecedented capacity of transformers to extract and integrate information from large and diverse data modalities, combined with the ever-increasing availability of biological and medical data, has the potential to revolutionize healthcare, promoting healthy longevity and mitigating the societal and economic impacts of global aging.

ParaCPI: A Parallel Graph Convolutional Network for Compound-Protein Interaction Prediction.

Identifying compound-protein interactions (CPIs) is critical in drug discovery, as accurate prediction of CPIs can remarkably reduce the time and cost of new drug development. The rapid growth of existing biological knowledge has opened up possibilities for leveraging known biological knowledge to predict unknown CPIs. However, existing CPI prediction models still fall short of meeting the needs of practical drug discovery applications. A novel parallel graph convolutional network model for CPI prediction (ParaCPI) is proposed in this study. This model constructs feature representation of compounds using a unique approach to predict unknown CPIs from known CPI data more effectively. Experiments are conducted on five public datasets, and the results are compared with current state-of-the-art (SOTA) models under three different experimental settings to evaluate the model's performance. In the three cold-start settings, ParaCPI achieves an average performance gain of 26.75%, 23.84%, and 14.68% in terms of area under the curve compared with the other SOTA models. In addition, the results of the experiments in the case study show ParaCPI's superior ability to predict unknown CPIs based on known data, with higher accuracy and stronger generalization compared with the SOTA models. Researchers can leverage ParaCPI to accelerate the drug discovery process.

Illuminating IT Landscapes: A Comparative Analysis of Discovery in ServiceNow and BMC Helix

Abstract: This research paper presents a comprehensive comparison of the Discovery applications offered by ServiceNow and BMC Helix. The study examines key features, discovery mechanisms, data collection capabilities, integration with CMDB, automation features, and analytics tools of both solutions. By evaluating their strengths and weaknesses in discovering and mapping IT assets and relationships, this paper aims to provide insights for organizations seeking to enhance their visibility into complex IT environments and improve overall IT service management.

FoodAtlas: Automated knowledge extraction of food and chemicals from literature

Automated generation of knowledge graphs that accurately capture published information can help with knowledge organization and access, which have the potential to accelerate discovery and innovation. Here, we present an integrated pipeline to construct a large-scale knowledge graph using large language models in an active learning setting. We apply our pipeline to the association of raw food, ingredients, and chemicals, a domain that lacks such knowledge resources. By using an iterative active learning approach of 4120 manually curated premise-hypothesis pairs as training data for ten consecutive cycles, the entailment model extracted 230,848 food-chemical composition relationships from 155,260 scientific papers, with 106,082 (46.0 %) of them never been reported in any published database. To augment the knowledge incorporated in the knowledge graph, we further incorporated information from 5 external databases and ontology sources. We then applied a link prediction model to identify putative food-chemical relationships that were not part of the constructed knowledge graph. Validation of the 443 hypotheses generated by the link prediction model resulted in 355 new food-chemical relationships, while results show that the model score correlates well (R2 = 0.70) with the probability of a novel finding. This work demonstrates how automated learning from literature at scale can accelerate discovery and support practical applications through reproducible, evidence-based capture of latent interactions of diverse entities, such as food and chemicals.

Discovery, Characterization and Synthetic Application of a Promiscuous Nonheme Iron Biocatalyst with Dual Hydroxylase/Desaturase Activity.

Alpha-ketoglutarate-dependent dioxygenases (αKGDs) have recently emerged as useful biocatalysts for C-H oxidation and functionalization. In this work, we characterized a new αKGD from aculene biosynthesis, AneA, which displays broad promiscuity toward a number of substrates with different ring systems. Unexpectedly, AneA was found to be capable of both desaturation and hydroxylation and require an amino ester motif on its substrate for productive catalysis. Insights gathered from the biochemical characterization and substrate-activity profiling of AneA enabled the development of a chemoenzymatic strategy toward several complex sesquiterpenoids.

Breadth versus depth: whole transcriptome sequencing has reduced sensitivity for detection of clinically relevant fusions compared to RNA comprehensive genomic profiling.

While there is great potential for unbiased next-generation sequencing (NGS) approaches-eg, whole transcriptome sequencing (WTS)-for exploration, discovery, and clinical application in the realm of oncology, there are limitations that should be considered when relying on these methodologies for clinical decision making. When using WTS for the detection of clinically relevant gene fusions in tumor specimens, a key consideration is whether a limited coverage depth (approximately 30-50X) is sufficient for detecting these events, especially in samples with low tumor purity. We demonstrate the reduced sensitivity of both a commercial WTS assay for the detection of clinically relevant fusions in analytical validation control samples and of a research use only (RUO) WTS assay for the detection of clinically relevant fusions in real-world clinical samples compared to RNA comprehensive genomic profiling (CGP). Notably, the RUO WTS assay would not have reported 30% (6/20) of fusions detected using RNA CGP assays in fusion-positive tumor samples, highlighting a potential disadvantage of broader sequencing.

Real-Time Wireless Sensing of Cardiomyocyte Contractility by Integrating Magnetic Microbeam and Oriented Nanofibers.

In vitro cardiomyocyte mechano-sensing platform is crucial for evaluating the mechanical performance of cardiac tissues and will be an indispensable tool for application in drug discovery and disease mechanism study. Magnetic sensing offers significant advantages in real-time, in situ wireless monitoring and resistance to ion interference. However, due to the mismatch between the stiffness of traditional rigid magnetic material and myocardial tissue, sensitivity is insufficient and it is difficult to achieve cell structure induction and three-dimensional cultivation. Herein, a magnetic sensing platform that integrates a neodymium-iron-boron/polydimethylsiloxane (NdFeB/PDMS) flexible microbeam with suspended and ordered polycaprolactone (PCL) nanofiber membranes was developed, providing a three-dimensional anisotropic culture environment for cardiomyocyte growth and simultaneously realizing in situ wireless contractility monitoring. The as-prepared sensor presented an ultrahigh sensitivity of 442.2 μV/μm and a deflection resolution of 2 μm. By continuously monitoring the cardiomyocyte growth status and drug stimulation feedback, we verified the capability of the platform to capture dynamic changes in cardiomyocyte contractility. This platform provides a perspective tool for evaluating cardiomyocyte maturity and drug performance.

Hyperconjugation‐Driven Isodesmic Reaction of Indoles and Anilines: Reaction Discovery, Mechanism Study, and Antitumor Application

Hyperconjugation‐Driven Isodesmic Reaction of Indoles and Anilines: Reaction Discovery, Mechanism Study, and Antitumor Application

Peptide and Protein Cysteine Modification Enabled by Hydrosulfuration of Ynamide.

Efficient functionalization of peptides and proteins has widespread applications in chemical biology and drug discovery. However, the chemoselective and site-selective modification of proteins remains a daunting task. Herein, a highly efficient chemo-, regio-, and stereoselective hydrosulfuration of ynamide was identified as an efficient method for the precise modification of peptides and proteins by uniquely targeting the thiol group of cysteine (Cys) residues. This novel method could be facilely operated in aqueous buffer and was fully compatible with a wide range of proteins, including small model proteins and large full-length antibodies, without compromising their integrity and functions. Importantly, this reaction provides the Z-isomer of the corresponding conjugates exclusively with superior stability, offering a precise approach to peptide and protein therapeutics. The potential application of this method in peptide and protein chemical biology was further exemplified by Cys-bioconjugation with a variety of ynamide-bearing functional molecules such as small molecule drugs, fluorescent/affinity tags, and PEG polymers. It also proved efficient in redox proteomic analysis through Cys-alkenylation. Overall, this study provides a novel bioorthogonal tool for Cys-specific functionalization, which will find broad applications in the synthesis of peptide/protein conjugates.

Deep dive into RNA: a systematic literature review on RNA structure prediction using machine learning methods

The discovery of non-coding RNAs (ncRNAs) has expanded our comprehension of RNAs’ inherent nature and capabilities. The intricate three-dimensional structures assumed by RNAs dictate their specific functions and molecular interactions. However, the limited number of mapped structures, partly due to experimental constraints of methods such as nuclear magnetic resonance (NMR), highlights the importance of in silico prediction solutions. This is particularly crucial in potential applications in therapeutic drug discovery. In this context, machine learning (ML) methods have emerged as prominent candidates, having previously demonstrated prowess in solving complex challenges across various domains. This review focuses on analyzing the development of ML-based solutions for RNA structure prediction, specifically oriented toward recent advancements in the deep learning (DL) domain. A systematic analysis of 33 works reveals insights into the representation of RNA structures, secondary structure motifs, and tertiary interactions. The review highlights current trends in ML methods used for RNA structure prediction, demonstrates the growing research involvement in this field, and summarizes the most valuable findings.