Protein-ligand interaction (PLI) is a critical step for drug discovery. Recently, protein pretrained language models (PLMs) have showcased exceptional performance across a wide range of protein-related tasks. However, a significant heterogeneity exists between the PLM and PLI tasks, leading to a degree of uncertainty. In this study, we propose a method that quantitatively assesses the significance of protein PLMs in PLI prediction. Specifically, we analyze the performance of three widely-used protein PLMs (TAPE, ESM-1b, and ProtTrans) on three PLI tasks (PDBbind, Kinase, and DUD-E). The model with pre-training consistently achieves improved performance and decreased time cost, demonstrating that enhance both the accuracy and efficiency of PLI prediction. By quantitatively assessing the transferability, the optimal PLM for each PLI task is identified without the need for costly transfer experiments. Additionally, we examine the contributions of PLMs on the distribution of feature space, highlighting the improved discriminability after pre-training. Our findings provide insights into the mechanisms underlying PLMs in PLI prediction and pave the way for the design of more interpretable and accurate PLMs in the future. Code and data are freely available at https://github.com/brian-zZZ/PLM-PLI.

With the increasing availability of large-scale QSAR (Quantitative Structure-Activity Relationship) datasets, collaborative analysis has become a promising approach for drug discovery. Traditional centralized analysis which typically concentrates data on a central server for training faces challenges such as data privacy and security. Distributed analysis such as federated learning offers a solution by enabling collaborative model training without sharing raw data. However, it may fail when the training data in the local devices are non-independent and identically distributed (non-IID). In this paper, we propose a novel framework for collaborative drug discovery using federated learning on non-IID datasets. We address the difficulty of training on non-IID data by globally sharing a small subset of data among all institutions. Our framework allows multiple institutions to jointly train a robust predictive model while preserving the privacy of their individual data. We leverage the federated learning paradigm to distribute the model training process across local devices, eliminating the need for data exchange. The experimental results on 15 benchmark datasets demonstrate that the proposed method achieves competitive predictive accuracy to centralized analysis while respecting data privacy. Moreover, our framework offers benefits such as reduced data transmission and enhanced scalability, making it suitable for large-scale collaborative drug discovery efforts.

The development of compounds that can selectively bind with non-canonical DNA structures has expanded in recent years. Junction DNA, including three-way junctions (3WJs) and four-way Holliday junctions (HJs), offer an intriguing target for developmental therapeutics as both 3WJs and HJs are involved in DNA replication and repair processes. However, there are a limited number of assays available for the analysis of junction DNA binding. Here, we describe the design and execution of multiplex fluorescent polyacrylamide gel electrophoresis (PAGE) and microscale thermophoresis (MST) assays that enable evaluation of junction-binding compounds. Two well characterised junction-binding compounds-a C6 linked bis-acridine ligand and an iron(II)-bound peptide helicate, which recognise HJs and 3WJs, respectively-were employed as probes for both MST and PAGE experiments. The multiplex PAGE assay expands beyond previously reported fluorescent PAGE as it combines four individual fluorophores that can be combined to visualise single-strands, pseudo-duplexes, and junction DNA present during 3WJ and HJ formation. The use of MST to identify the binding affinity of junction binding agents is, to our knowledge, first reported example of this technique. The combined use of PAGE and MST provides complementary results for the visualisation of 3WJ and HJ formation and the direct binding affinity (Kd and EC50) of these agents. These assays can be used to aid the discovery and design of new therapeutics targeting non-canonical nucleic acid structures.

Use of cationic lipid vesicles (liposomes) can yield large amounts of nucleic acid entrapped inside the vesicles and/or bound to the external surface of the vesicles. To show a method to prepare asymmetric lipid vesicles (liposomes) with high amounts of entrapped nucleic acid is possible, symmetric and asymmetric lipid vesicles composed of mixtures of neutral (zwitterionic), anionic, and/or cationic phospholipids were formed in the presence of oligo DNA. For symmetric large unilamellar vesicles nucleic acid association with vesicles was roughly 100 times greater for vesicles with a net cationic charge than for vesicles having a net neutral or anionic net charge. A high degree of association between nucleic acid and lipid was also achieved using asymmetric large unilamellar vesicles with a net cationic charge in their inner leaflet, even when they had an anionic charge in their outer leaflet. In contrast, asymmetric vesicles in which only the outer leaflet had a net cationic charge had only low amounts of vesicle-associated nucleic acid, similar in amount to the amount of nucleic acid associated with asymmetric vesicles with an outer leaflet having a net anionic charge. These results indicate that in asymmetric vesicles with cationic lipid enriched inner leaflets nucleic acid is largely entrapped inside the vesicle lumen rather than bound to their external surface, and that asymmetric vesicles can be used to trap high amounts of nucleic acid even when using a lipid composition in the outer leaflet of a lipid vesicle that does not associate with nucleic acids. Such asymmetrically charged vesicles should have applications in studies of membrane protein-nucleic acid interactions as well as in studies of how membrane charge asymmetry can influence membrane protein structure, orientation, and function.

Long non-coding RNA (lncRNA) are shown to be closely associated with cancer metastatic events (CME, e.g., cancer cell invasion, intravasation, extravasation, proliferation) that collaboratively accelerate malignant cancer spread and cause high mortality rate in patients. Clinical trials may accurately uncover the relationships between lncRNAs and CMEs; however, it is time-consuming and expensive. With the accumulation of data, there is an urgent need to find efficient ways to identify these relationships. Herein, a graph embedding representation-based predictor (VGEA-LCME) for exploring latent lncRNA-CME associations is introduced. In VGEA-LCME, a heterogeneous combined network is constructed by integrating similarity and linkage matrix that can maintain internal and external characteristics of networks, and a variational graph auto-encoder serves as a feature generator to represent arbitrary lncRNA and CME pair. The final robustness predicted result is obtained by ensemble classifier strategy via cross-validation. Experimental comparisons and literature verification show better remarkable performance of VGEA-LCME, although the similarities between CMEs are challenging to calculate. In addition, VGEA-LCME can further identify organ-specific CMEs. To the best of our knowledge, this is the first computational attempt to discover the potential relationships between lncRNAs and CMEs. It may provide support and new insight for guiding experimental research of metastatic cancers. The source code and data are available at https://github.com/zhuyuan-cug/VGAE-LCME.

Liver is an important metabolic organ in human body and is sensitive to toxic chemicals or drugs. Adverse reactions caused by drug hepatotoxicity will damage the liver and hepatotoxicity is the leading cause of removal of approved drugs from the market. Therefore, it is of great significance to identify liver toxicity as early as possible in the drug development process. In this study, we developed a predictive model for drug hepatotoxicity based on histopathological whole silde images (WSI) which are the by-product of drug experiments and have received little attention. To better represent the WSIs, we constructed a graph representation for each WSI by dividing it into small patches, taking sampled patches as nodes and calculating the correlation coefficients between node features as the edges of the graph structure. Then a WSI-level graph convolutional network (GCN) was built to effectively extract the node information of the graph and predict the toxicity. In addition, we introduced a gated attention global context vector (gaGCV) to combine the global context to make node features to contain more comprehensive information. The results validated on rat liver in vivo data from the Open TG-GATES show that the use of WSI for the prediction of toxicity is feasible and effective.

Deep learning is improving and changing the process of de novo molecular design at a rapid pace. In recent years, great progress has been made in drug discovery and development by using deep generative models for de novo molecular design. However, most of the existing methods are string-based or graph-based and are limited by the lack of some very important properties, such as the three-dimensional information of molecules. We propose DNMG, a deep generative adversarial network (GAN) combined with transfer learning. Specifically, we use a Wasserstein-variant GAN based network architecture that considers the 3D grid spatial information of the ligand with atomic physicochemical properties to generate a representation of the molecule, which is then parsed into SMILES strings using an improved captioning network. Comprehensive in experiments demonstrate the ability of DNMG to generate valid and novel drug-like ligands. The DNMG model is used to design inhibitors for three targets, MK14, FNTA, and CDK2. The computational results show that the molecules generated by DNMG have better binding ability to the target proteins and better physicochemical properties. Overall, our deep generative model has excellent potential to generate molecules with high binding affinity for targets and explore the space of drug-like chemistry.

The enhancer is a DNA sequence that can increase the activity of promoters and thus speed up the frequency of gene transcription. The enhancer plays an essential role in activating gene expression. Currently, gene sequencing technology has been developed for 30 years from the first generation to the third generation, and a variety of biological sequence data have increased significantly every year. Due to the importance of enhancer functions, it is very expensive to identify enhancers through biochemical experiments. Therefore, we need to study new methods for the identification and classification of enhancers. Based on the K-mer principle this study proposed a feature extraction method that others have not used in convolutional neural networks. Then, we combined it with one-hot encoding to build an efficient one-dimensional convolutional neural network ensemble model for predicting enhancers and their strengths. Finally, we used five commonly used classification problem evaluation indicators to compare with the models proposed by other researchers. The model proposed in this paper has a better performance by using the same independent test dataset as other models.

The Shwachman-Diamond syndrome (SDS) is a rare inherited ribosomopathy that is predominantly caused by mutations in the Shwachman-Bodian-Diamond Syndrome gene (SBDS). SBDS is a ribosomal maturation factor that is essential for the release of eukaryotic translation initiation factor 6 (eIF6) from the 60S ribosomal subunits during the late stages of 60S maturation. Release of eIF6 is critical to permit inter-subunit interactions between the 60S and 40S subunits and to form translationally competent 80S monosomes. SBDS has three key domains that are highly flexible and adopt varied conformations in solution. To better understand the domain dynamics of SBDS upon binding to 60S and to assess the effects of SDS-disease specific mutations, we aimed to site-specifically label individual domains of SBDS. Here we detail the generation of a fluorescently labeled SBDS to monitor the dynamics of select domains upon binding to 60S. We describe the incorporation of 4-azido-l-phenylalanine (4AZP), a noncanonical amino acid in human SBDS. Site-specific labeling of SBDS using fluorophore and assessment of 60S binding activity are also described. Such labeling approaches to capture the interactions of individual domains of SBDS with 60S is also applicable to study the dynamics of other multi-domain proteins that interact with the ribosomal subunits.

Self-supervised learning has shown superior performance on graph-related tasks in recent years. The most advanced methods are based on contrast learning, which severely limited by structured data augmentation techniques and complex training methods. Generative self-supervised learning, especially graph autoencoders (GAEs), can prevent the above dependence and has been demonstrated as an effective approach. In addition, most previous works only reconstruct the graph topological structure or node features. Few works consider both and combine them together to obtain their complementary information. To overcome these problems, we propose a generative self-supervised graph representation learning methodology named Multi-View Dual-decoder Graph Autoencoder (MDGA). Specifically, we first design a multi-sample graph learning strategy which benefits the generalization of the dual-decoder graph autoencoder. Moreover, the proposed model reconstructs the graph topological structure with a traditional GAE and extracts node attributes by masked feature reconstruction. Experimental results on five public benchmark datasets demonstrate that MDGA outperforms state-of-the-art methods in both node classification and link prediction tasks.

The nucleosome is the fundamental building block of chromatin. Changes taking place at the nucleosome level are the molecular basis of chromatin transactions with various enzymes and factors. These changes are directly and indirectly regulated by chromatin modifications such as DNA methylation and histone post-translational modifications including acetylation, methylation, and ubiquitylation. Nucleosomal changes are often stochastic, unsynchronized, and heterogeneous, making it very difficult to monitor with traditional ensemble averaging methods. Diverse single-molecule fluorescence approaches have been employed to investigate the structure and structural changes of the nucleosome in the context of its interactions with various enzymes such as RNA Polymerase II, histone chaperones, transcription factors, and chromatin remodelers. We utilize diverse single-molecule fluorescence methods to study the nucleosomal changes accompanying these processes, elucidate the kinetics of these processes, and eventually learn the implications of various chromatin modifications in directly regulating these processes. The methods include two- and three-color single-molecule fluorescence resonance energy transfer (FRET), single-molecule fluorescence correlation spectroscopy, and fluorescence (co-)localization. Here we report the details of the two- and three-color single-molecule FRET methods we currently use. This report will help researchers design their single-molecule FRET approaches to investigating chromatin regulation at the nucleosome level.

Recent advances in single molecule imaging have allowed the evolution of biochemical techniques that directly visualize protein-DNA interactions in real-time. These techniques rely on diffraction limited total internal reflection microscopy (TIRFM), and have significantly improved our understanding of RNA transcription, DNA replication, homologous recombination, and general DNA repair in the context of chromatin. Here we described a general single molecule TIRFM technique called DNA curtains to directly visualize how enzymes function on chromatinized DNA. The goal of this manuscript is to introduce the reader to methods to express and reconstitute nucleosomes on long stretches of DNA, and to directly visualize this process using DNA curtains with TIRFM.

Antibody drugs have become a key part of biotherapeutics. Patients suffering from various diseases have benefited from antibody therapies. However, its development process is rather long, expensive and risky. To speed up the process, reduce cost and improve success rate, artificial intelligence, especially deep learning methods, have been widely used in all aspects of preclinical antibody drug development, from library generation to hit identification, developability screening, lead selection and optimization. In this review, we systematically summarize antibody encodings, deep learning architectures and models used in preclinical antibody drug discovery and development. We also critically discuss challenges and opportunities, problems and possible solutions, current applications and future directions of deep learning in antibody drug development.

Heart rate variability (HRV) is an important indicator of autonomic nervous system activity and can be used for the identification of affective states. The development of remote Photoplethysmography (rPPG) technology has made it possible to measure pulse rate variability (PRV) using a camera without any sensor-skin contact, which is highly correlated to HRV, thus, enabling contactless assessment of emotional states. In this study, we employed ten machine learning techniques to identify emotions using camera-based PRV features. Our experimental results show that the best classification model achieved a coordination correlation coefficient of 0.34 for value recognition and 0.36 for arousal recognition. The rPPG-based measurement has demonstrated promising results in detecting HAHV (high-arousal high-valence) emotions with high accuracy. Furthermore, for emotions with less noticeable variations, such as sadness, the rPPG-based measure outperformed the baseline deep network for facial expression analysis.

CONTEXT: Surface electromyography (sEMG) signals contain rich information recorded from muscle movements and therefore reflect the user's intention. sEMG has seen dominant applications in rehabilitation, clinical diagnosis as well as human engineering, etc. However, current feature extraction methods for sEMG signals have been seriously limited by their stochasticity, transiency, and non-stationarity. OBJECTIVE: Our objective is to combat the difficulties induced by the aforementioned downsides of sEMG and thereby extract representative features for various downstream movement recognition. METHOD: We propose a novel 3-axis view of sEMG features composed of temporal, spatial, and channel-wise summary. We leverage the state-of-the-art architecture Transformer to enforce efficient parallel search and to get rid of limitations imposed by previous work in gesture classification. The transformer model is designed on top of an attention-based module, which allows for the extraction of global contextual relevance among channels and the use of this relevance for sEMG recognition. RESULTS: We compared the proposed method against existing methods on two Ninapro datasets consisting of data from both healthy people and amputees. Experimental results show the proposed method attains the state-of-the-art (SOTA) accuracy on both datasets. We further show that the proposed method enjoys strong generalization ability: a new SOTA is achieved by pretraining the model on a different dataset followed by fine-tuning it on the target dataset.

Advances in Nuclear Magnetic Resonance (NMR) spectroscopy have allowed for the identification and characterization of movements in enzymes over the last 20 years that has also revealed the complexities of allosteric coupling. For example, many of the inherent movements of enzymes, and proteins in general, have been shown to be highly localized but nonetheless still coupled over long distances. Such partial couplings provide challenges to both identifying allosteric networks of dynamic communication and determining their roles in catalytic function. We have developed an approach to help identify and engineer enzyme function, called Relaxation And Single Site Multiple Mutations (RASSMM). This approach is a powerful extension of mutagenesis and NMR that is based on the observation that multiple mutations to a single site distal to the active site allosterically induces different effects to networks. Such an approach generates a panel of mutations that can also be subjected to functional studies in order to match catalytic effects with changes to coupled networks. In this review, the RASSMM approach is briefly outlined together with two applications that include cyclophilin-A and Biliverdin Reductase B.

The implementation of early cancer detection benefits the treatment outcomes with remarkably improved survival rate through the detection of rare circulating biomarkers in body fluids. Spectroscopic technologies play a crucial role in sensitive biomarker measurements by outputting extremely strong signals. In particular, the aggregation enhanced fluorescence and Raman technologies feature the detection of targets down to single-molecule level, thereby demonstrating the great promise of early cancer detection. In this review, we focus on the aggregation-induced emission (AIE) and aggregation-related surface-enhanced Raman scattering (SERS) spectroscopic strategies for detecting cancer biomarkers. We discuss the AIE and SERS based biomarker detection using target-driven aggregation as well as the aggregated nanoprobes. Furthermore, we deliberate on the progress of developing AIE and SERS integrated platforms. Ultimately, we put forth the potential challenges and perspectives on the way to use these two spectroscopic technologies in clinical settings. It is expected this review can inspire the design of AIE and SERS integrated platform for highly sensitive and accurate cancer detection.

CONTEXT: Novel kinds of antibiotics are needed to combat the emergence of antibacterial resistance. Natural products (NPs) have shown potential as antibiotic candidates. Current experimental methods are not yet capable of exploring the massive, redundant, and noise-involved chemical space of NPs. In silico approaches are needed to select NPs as antibiotic candidates. OBJECTIVE: This study screens out NPs with antibacterial efficacy guided by both TCM and modern medicine and constructed a dataset aiming to serve the new antibiotic design. METHOD: A knowledge-based network is proposed in this study involving NPs, herbs, the concepts of TCM, and the treatment protocols (or etiologies) of infectious in modern medicine. Using this network, the NPs candidates are screened out and compose the dataset. Feature selection of machine learning approaches is conducted to evaluate the constructed dataset and statistically validate the im- portance of all NPs candidates for different antibiotics by a classification task. RESULTS: The extensive experiments prove the constructed dataset reaches a convincing classification performance with a 0.9421 weighted accuracy, 0.9324 recall, and 0.9409 precision. The further visu- alizations of sample importance prove the comprehensive evaluation for model interpretation based on medical value considerations.

DNA microarrays are very useful tools to study the realm of nucleic acids interactions at high throughput. The conventional approach to microarray synthesis employs phosphoramidite chemistry and yields unmodified DNA generally attached to a surface at the 3' terminus. Having a freely accessible 3'-OH instead of 5'-OH is desirable too, and being able to introduce nucleoside analogs in a combinatorial manner is highly relevant in the context of nucleic acid therapeutics and in aptamer research. Here, we describe an enzymatic approach to the synthesis of high-density DNA microarrays that can also contain chemical modifications. The method uses a standard DNA microarray, to which a DNA primer is covalently bound through photocrosslinking. The extension of the primer with a DNA polymerase yields double-stranded DNA but is also amenable to the incorporation of modified dNTPs. Further processing with T7 exonuclease, which catalyzes the degradation of DNA in a specific (5'→3') direction, results in template strand removal. Overall, the method produces surface-bound natural and non-natural DNA oligonucleotides, is applicable to commercial microarrays and paves the way for the preparation of combinatorial, chemically modified aptamer libraries.

As an important task of natural language processing, medication recommendation aims to recommend medication combinations according to the electronic health record, which can also be regarded as a multi-label classification task. But patients often have multiple diseases simultaneously, and the model must consider drug-drug interactions (DDI) of medication combinations when recommending medications, making medication recommendation more difficult. There is little existing work to explore the changes in patient conditions. However, these changes may point to future trends in patient conditions that are critical for reducing DDI rates in recommended drug combinations. In this paper, we proposed the Patient Information Mining Network (PIMNet), which models the current core medications of patient by mining the temporal and spatial changes of patient medication order and patient condition vector, and allocates some auxiliary medications as the currently recommended medication combination. The experimental results show that the proposed model greatly reduces the recommended DDI of medications while achieving results no lower than the state-of-the-art results.