KinomeMETA: meta-learning enhanced kinome-wide polypharmacology profiling.

Kinase inhibitors are crucial in cancer treatment, but drug resistance and side effects hinder the development of effective drugs. To address these challenges, it is essential to analyze the polypharmacology of kinase inhibitor and identify compound with high selectivity profile. This study presents KinomeMETA, a framework for profiling the activity of small molecule kinase inhibitors across a panel of 661 kinases. By training a meta-learner based on a graph neural network and fine-tuning it to create kinase-specific learners, KinomeMETA outperforms benchmark multi-task models and other kinase profiling models. It provides higher accuracy for understudied kinases with limited known data and broader coverage of kinase types, including important mutant kinases. Case studies on the discovery of new scaffold inhibitors for membrane-associated tyrosine- and threonine-specific cdc2-inhibitory kinase and selective inhibitors for fibroblast growth factor receptors demonstrate the role of KinomeMETA in virtual screening and kinome-wide activity profiling. Overall, KinomeMETA has the potential to accelerate kinase drug discovery by more effectively exploring the kinase polypharmacology landscape.

The emergence of machine learning force fields in drug design.

In the field of molecular simulation for drug design, traditional molecular mechanic force fields and quantum chemical theories have been instrumental but limited in terms of scalability and computational efficiency. To overcome these limitations, machine learning force fields (MLFFs) have emerged as a powerful tool capable of balancing accuracy with efficiency. MLFFs rely on the relationship between molecular structures and potential energy, bypassing the need for a preconceived notion of interaction representations. Their accuracy depends on the machine learning models used, and the quality and volume of training data sets. With recent advances in equivariant neural networks and high-quality datasets, MLFFs have significantly improved their performance. This review explores MLFFs, emphasizing their potential in drug design. It elucidates MLFF principles, provides development and validation guidelines, and highlights successful MLFF implementations. It also addresses potential challenges in developing and applying MLFFs. The review concludes by illuminating the path ahead for MLFFs, outlining the challenges to be overcome and the opportunities to be harnessed. This inspires researchers to embrace MLFFs in their investigations as a new tool to perform molecular simulations in drug design.

Construction of an alanine dehydrogenase gene deletion strain for vaccine development against Nocardia seriolae in hybrid snakehead (Channa maculata ♀ × Channa argus ♂).

Nocardia seriolae is the main pathogen of fish nocardiosis. In our previous study, alanine dehydrogenase was identified as a potential virulence factor of N. seriolae. On the basis of this fact, the alanine dehydrogenase gene of N. seriolae (NsAld) was knocked out to establish the strain ΔNsAld for vaccine development against fish nocardiosis in this study. The LD50 of strain ΔNsAld was 3.90 × 105 CFU/fish, higher than that of wild strain (5.28 × 104 CFU/fish) significantly (p < 0.05). When the strain ΔNsAld was used as a live vaccine to immunize hybrid snakehead (Channa maculata ♀ × Channa argus ♂) at 2.47 × 105 CFU/fish by intraperitoneal injection, the non-specific immune indexes (LZM, CAT, AKP, ACP and SOD activities), specific antibody (IgM) titers and several immune-related genes (CD4, CD8α, IL-1ß, MHCIα, MHCIIα and TNFα) were up-regulated in different tissues, indicating that this vaccine could induce humoral and cell-mediated immune responses. Furthermore, the relative percentage survival (RPS) of ΔNsAld vaccine was calculated as 76.48% after wild N. seriolae challenge. All these results suggest that the strain ΔNsAld could be a potential candidate for live vaccine development to control fish nocardiosis in aquaculture.

Induction of attenuated Nocardia seriolae and their use as live vaccine trials against fish nocardiosis.

Nocardia seriolae, a Gram-positive facultative intercellular pathogen, has been identified as the causative agent of fish nocardiosis, causing substantial mortality and morbidity of a wide range of fish species. Looking into that fact, the effective vaccine against this pathogen is urgently needed to control the significant losses in aquaculture practices. In order to induct attenuated strains for developing the potential live vaccines, the mutagenic N. seriolae strain S-250 and U-20 were obtained from wild-type strain ZJ0503 through continuous passaging and ultraviolet (UV) irradiation, respectively. Additionally, the biological characteristic, virulence, stability, mediating immune response and supplying protective efficacy to hybrid snakehead of the S-250 and U-20 strains were determined in the present study. The results showed that U-20 strain displayed dramatic changes in morphological characteristic and significant decreased in the virulence to hybrid snakehead, while that of S-250 strain had no obvious different in comparison to ZJ0503 strain. When hybrid snakehead were intraperitoneally injected with ZJ0503, S-250 and U-20 strains at their respective sub-clinical dosage, the non-specific immunity parameters (serum LYZ, POD, ACP, AKP and SOD activities), specific antibody (IgM) titers production and immune-related genes (CC1, CC2, IL-1ß, IL-8, TNFα, IFNγ, MHCIα, MHCIIα, CD4, CD8α, TCRα and TCRß) expression were up-regulated, indicating that they were able to trigger humoral and cell-mediated immune responses. Furthermore, the protective efficacy in hybrid snakehead after vaccination with ZJ0503, S-250 and U-20 strains, in terms of relative percentage survival (RPS), were 28.85%, 56.89% and 89.65% respectively. Taken together, two attenuated N. seriolae strains S-250 and U-20 were obtained successfully and they could elicit strong immune response and supply protective efficacy to hybrid snakehead against N. seriolae, which suggested that these two attenuated strains were the potential candidates for live vaccine development to control fish nocardiosis in aquaculture.

Identification of a secretory heme-binding protein from Nocardia seriolae involved in cell apoptosis.

According to the whole-genome bioinformatics analysis, a heme-binding protein from Nocardia seriolae (HBP) was found. HBP was predicted to be a bacterial secretory protein, located at mitochondrial membrane in eukaryotic cells and have a similar protein structure with the heme-binding protein of Mycobacterium tuberculosis, Rv0203. In this study, HBP was found to be a secretory protein and co-localized with mitochondria in FHM cells. Quantitative analysis of mitochondrial membrane potential value, caspase-3 activity, and transcription level of apoptosis-related genes suggested that overexpression of HBP protein can induce cell apoptosis. In conclusion, HBP was a secretory protein which may target to mitochondria and involve in cell apoptosis in host cells. This research will promote the function study of HBP and deepen the comprehension of the virulence factors and pathogenic mechanisms of N. seriolae.

Oxygen-functionalized TlTe buckled honeycomb from first-principles study.

A sizable band gap is crucial for the applications of topological insulators at room temperature. By first-principles calculations, we found that oxygen-functionalized TlTe buckled honeycomb, namely TlTeO, possessed quantum spin Hall (QSH) state with a sizable band gap of 0.17 eV, which owns potential applications at the room temperature. The QSH phase of TlTeO arose from the SOC-induced p-p band gap opening. In addition, the QSH phase was further confirmed by the topological invariant Z2 and gapless edge state in the bulk gap. Significantly, the QSH phase is robustly against the external strain and possesses more than 75% oxygen coverage, making the QSH effect of TlTeO easy to be achieved experimentally. Thus, the oxygen-functionalized TlTeO film is a fine candidate material for the topological device design and fabrication.

A novel method to detect Listeria monocytogenes via superparamagnetic lateral flow immunoassay.

A novel strip test system combining immunomagnetic separation with lateral flow immunoassay (LFIA) was established for the accurate detection of Listeria monocytogenes. In this system, a pair of matched monoclonal antibodies was used to construct a sandwich immunoassay, in which superparamagnetic particles were coupled with one of the antibodies as a labeled antibody to capture the target bacteria, while the other antibody was immobilized on the detection zone. After a 20-min reaction, the strips were analyzed by a novel instrument which could detect the magnetic signal of the immunocomplex in a magnetic field. Sensitivity evaluation showed that the limit of detection (LOD) of the superparamagnetic LFIA system for L. monocytogenes was 10(4) CFU/mL, which was at least one log lower than conventional LFIA. No cross-reaction was observed when Salmonella, Escherichia coli O157:H7, or three types of harmless Listeria strains were tested. Further evaluation with actual food samples indicated that the superparamagnetic LFIA system showed 100 % concordance with real-time PCR. Therefore, this novel superparamagnetic LFIA system could be used as a rapid, sensitive, and specific method for the detection of L. monocytogenes.

Cosine convolutional neural network and its application for seizure detection.

Traditional convolutional neural networks (CNNs) often suffer from high memory consumption and redundancy in their kernel representations, leading to overfitting problems and limiting their application in real-time, low-power scenarios such as seizure detection systems. In this work, a novel cosine convolutional neural network (CosCNN), which replaces traditional kernels with the robust cosine kernel modulated by only two learnable factors, is presented, and its effectiveness is validated on the tasks of seizure detection. Meanwhile, based on the cosine lookup table and KL-divergence, an effective post-training quantization algorithm is proposed for CosCNN hardware implementation. With quantization, CosCNN can achieve a nearly 75% reduction in the memory cost with almost no accuracy loss. Moreover, we design a configurable cosine convolution accelerator on Field Programmable Gate Array (FPGA) and deploy the quantized CosCNN on Zedboard, proving the proposed seizure detection system can operate in real-time and low-power scenarios. Extensive experiments and comparisons were conducted using two publicly available epileptic EEG databases, the Bonn database and the CHB-MIT database. The results highlight the performance superiority of the CosCNN over traditional CNNs as well as other seizure detection methods.

Epileptic Seizure Detection with an End-to-End Temporal Convolutional Network and Bidirectional Long Short-Term Memory Model.

Automatic seizure detection plays a key role in assisting clinicians for rapid diagnosis and treatment of epilepsy. In view of the parallelism of temporal convolutional network (TCN) and the capability of bidirectional long short-term memory (BiLSTM) in mining the long-range dependency of multi-channel time-series, we propose an automatic seizure detection method with a novel end-to-end TCN-BiLSTM model in this work. First, raw EEG is filtered with a 0.5-45 Hz band-pass filter, and the filtered data are input into the proposed TCN-BiLSTM network for feature extraction and classification. Post-processing process including moving average filtering, thresholding and collar technique is then employed to further improve the detection performance. The method was evaluated on two EEG database. On the CHB-MIT scalp EEG database, our method achieved a segment-based sensitivity of 94.31%, specificity of 97.13%, and accuracy of 97.09%. Meanwhile, an event-based sensitivity of 96.48% and an average false detection rate (FDR) of 0.38/h were obtained. On the SH-SDU database we collected, the segment-based sensitivity of 94.99%, specificity of 93.25%, and accuracy of 93.27% were achieved. In addition, an event-based sensitivity of 99.35% and a false detection rate of 0.54/h were yielded. The total detection time consumed for 1[Formula: see text]h EEG data was 5.65[Formula: see text]s. These results demonstrate the superiority and promising potential of the proposed method in real-time monitoring of epileptic seizures.

Deep representation learning of chemical-induced transcriptional profile for phenotype-based drug discovery.

Artificial intelligence transforms drug discovery, with phenotype-based approaches emerging as a promising alternative to target-based methods, overcoming limitations like lack of well-defined targets. While chemical-induced transcriptional profiles offer a comprehensive view of drug mechanisms, inherent noise often obscures the true signal, hindering their potential for meaningful insights. Here, we highlight the development of TranSiGen, a deep generative model employing self-supervised representation learning. TranSiGen analyzes basal cell gene expression and molecular structures to reconstruct chemical-induced transcriptional profiles with high accuracy. By capturing both cellular and compound information, TranSiGen-derived representations demonstrate efficacy in diverse downstream tasks like ligand-based virtual screening, drug response prediction, and phenotype-based drug repurposing. Notably, in vitro validation of TranSiGen's application in pancreatic cancer drug discovery highlights its potential for identifying effective compounds. We envisage that integrating TranSiGen into the drug discovery and mechanism research holds significant promise for advancing biomedicine.

Self-attention-based convolutional neural network and time-frequency common spatial pattern for enhanced motor imagery classification.

BACKGROUND: Motor imagery (MI) based brain-computer interfaces (BCIs) have promising potentials in the field of neuro-rehabilitation. However, due to individual variations in active brain regions during MI tasks, the challenge of decoding MI EEG signals necessitates improved classification performance for practical application. NEW METHOD: This study proposes a self-attention-based Convolutional Neural Network (CNN) in conjunction with a time-frequency common spatial pattern (TFCSP) for enhanced MI classification. Due to the limited availability of training data, a data augmentation strategy is employed to expand the scale of MI EEG datasets. The self-attention-based CNN is trained to automatically extract the temporal and spatial information from EEG signals, allowing the self-attention module to select active channels by calculating EEG channel weights. TFCSP is further implemented to extract multiscale time-frequency-space features from EEG data. Finally, the EEG features derived from TFCSP are concatenated with those from the self-attention-based CNN for MI classification. RESULTS: The proposed method is evaluated on two publicly accessible datasets, BCI Competition IV IIa and BCI Competition III IIIa, yielding mean accuracies of 79.28 % and 86.39 %, respectively. CONCLUSIONS: Compared with state-of-the-art methods, our approach achieves superior classification results in accuracy. Self-attention-based CNN combining with TFCSP can make full use of the time-frequency-space information of EEG, and enhance the classification performance.

Long-term consumption of caffeine-free high sucrose cola beverages aggravates the pathogenesis of EAE in mice.

Epidemiological data provide strong evidence of dramatically increasing incidences of many autoimmune diseases in the past few decades, mainly in western and westernized countries. Recent studies clearly revealed that 'Western diet' increases the risk of autoimmune diseases at least partially via disrupting intestinal tight junctions and altering the construction and metabolites of microbiota. However, the role of high sucrose cola beverages (HSCBs), which are one of the main sources of added sugar in the western diet, is barely known. Recently, a population study showed that regular consumption of sugar-sweetened beverages is associated with increased risk of seropositive rheumatoid arthritis in women, which provokes interest in the genuine effects of these beverages on the pathogenesis of autoimmune diseases and the underlying mechanisms. Here we showed that long-term consumption of caffeine-free HSCBs aggravated the pathogenesis of experimental autoimmune encephalomyelitis in mice in a microbiota-dependent manner. Further investigation revealed that HSCBs altered community structure of microbiota and increased Th17 cells. High sucrose consumption had similar detrimental effects while caffeine contamination limited the infiltrated pathogenic immune cells and counteracted these effects. These results uncovered a deleterious role of decaffeinated HSCBs in aggravating the pathogenesis of experimental autoimmune encephalomyelitis in mice.

[Rapid detection of Listeria monocytogenes by immunomagnetic separation combined with selective medium].

Listeria monocytogenes is a pathogenic bacterium, therefore, it is essential for food safety monitoring to establish a rapid and specific detecting method. In this study, immunomagnetic beads and selective medium were combined to detect Listeria monocytogenes at different concentrations (10(1)-10(5) CFU/mL). Other three types of Listeria spp., Staphylococcus aureus and Vibrio parahaemolyticus were also detected to conduct the cross-reaction analysis. Meanwhile, contaminated milk samples were prepared to explore the limit of detection of immunomagnetic beads combining with selective medium. Results showed that Listeria monocytogenes with the concentration of 10(3) CFU/mL and above was successfully detected. Milk samples were detected within 6 hours, with a detection limit of 0.7 CFU/mL. The method developed is capable of detecting milk samples within 30 h, which is 38 h faster compared with national standard method with the same sensitivity.