odd skipped-related 2 as a novel mark for labeling the proximal convoluted tubule within the zebrafish kidney.

The proximal convoluted tubule (PCT) of the kidney is a crucial functional segment responsible for reabsorption, secretion, and the maintenance of electrolyte and water balance within the renal tubule. However, there is a lack of a well-defined endogenous transgenic line for studying PCT morphogenesis. By analyzing single-cell transcriptome data from the adult zebrafish kidney, we have identified the expression of odd-skipped-related 2 (osr2, which encodes an odd-skipped zinc-finger transcription factor) in the PCT. To gain insight into the role of osr2 in PCT morphogenesis, we have generated a transgenic zebrafish line Tg(osr2:EGFP), expressing enhanced green fluorescent protein (EGFP). The EGFP expression pattern closely mirrors that of endogenous Osr2, faithfully recapitulating its native expression profile. During kidney development, we can use EGFP to track PCT development, which is also preserved in adult zebrafish. Additionally, osr2:EGFP-labeled zebrafish PCT fragments displayed short lengths with infrequent overlap, rendering them conducive for nephrons counting. The generation of Tg(osr2:EGFP) transgenic line is accompanied by simultaneous disruption of osr2 activity. Importantly, our findings demonstrate that osr2 inactivation had no discernible impact on the development and regeneration of Tg(osr2:EGFP) zebrafish nephrons. Overall, the establishment of this transgenic zebrafish line offers a valuable tool for both genetic and chemical analysis of PCT.

Proenkephalin-A secreted by renal proximal tubules functions as a brake in kidney regeneration.

Organ regeneration necessitates precise coordination of accelerators and brakes to restore organ function. However, the mechanisms underlying this intricate molecular crosstalk remain elusive. In this study, the level of proenkephalin-A (PENK-A), expressed by renal proximal tubular epithelial cells, decreases significantly with the loss of renal proximal tubules and increased at the termination phase of zebrafish kidney regeneration. Notably, this change contrasts with the role of hydrogen peroxide (H2O2), which acts as an accelerator in kidney regeneration. Through experiments with penka mutants and pharmaceutical treatments, we demonstrate that PENK-A inhibits H2O2 production in a dose-dependent manner, suggesting its involvement in regulating the rate and termination of regeneration. Furthermore, H2O2 influences the expression of tcf21, a vital factor in the formation of renal progenitor cell aggregates, by remodeling H3K4me3 in renal cells. Overall, our findings highlight the regulatory role of PENK-A as a brake in kidney regeneration.

Characterization of the Difference between Day and Night Temperatures on the Growth, Photosynthesis, and Metabolite Accumulation of Tea Seedlings.

Currently, the effects of the differences between day and night temperatures (DIFs) on tea plant are poorly understood. In order to investigate the influence of DIFs on the growth, photosynthesis, and metabolite accumulation of tea plants, the plants were cultivated under 5 °C (25/20 °C, light/dark), 10 °C (25/15 °C, light/dark), and 15 °C (25/10 °C, light/dark). The results showed that the growth rate of the new shoots decreased with an increase in the DIFs. There was a downward trend in the photosynthesis among the treatments, as evidenced by the lowest net photosynthetic rate and total chlorophyll at a DIF of 15 °C. In addition, the DIFs significantly affected the primary and secondary metabolites. In particular, the 10 °C DIF treatment contained the lowest levels of soluble sugars, tea polyphenols, and catechins but was abundant in caffeine and amino acids, along with high expression levels of theanine synthetase (TS3) and glutamate synthase (GOGAT). Furthermore, the transcriptome data revealed that the differentially expressed genes were enriched in valine, leucine, and isoleucine degradation, flavone/flavonol biosyntheses, flavonoid biosynthesis, etc. Therefore, we concluded that a DIF of 10 °C was suitable for the protected cultivation of tea plants in terms of the growth and the quality of a favorable flavor of tea, which provided a scientific basis for the protected cultivation of tea seedlings.

MolFilterGAN: a progressively augmented generative adversarial network for triaging AI-designed molecules.

Artificial intelligence (AI)-based molecular design methods, especially deep generative models for generating novel molecule structures, have gratified our imagination to explore unknown chemical space without relying on brute-force exploration. However, whether designed by AI or human experts, the molecules need to be accessibly synthesized and biologically evaluated, and the trial-and-error process remains a resources-intensive endeavor. Therefore, AI-based drug design methods face a major challenge of how to prioritize the molecular structures with potential for subsequent drug development. This study indicates that common filtering approaches based on traditional screening metrics fail to differentiate AI-designed molecules. To address this issue, we propose a novel molecular filtering method, MolFilterGAN, based on a progressively augmented generative adversarial network. Comparative analysis shows that MolFilterGAN outperforms conventional screening approaches based on drug-likeness or synthetic ability metrics. Retrospective analysis of AI-designed discoidin domain receptor 1 (DDR1) inhibitors shows that MolFilterGAN significantly increases the efficiency of molecular triaging. Further evaluation of MolFilterGAN on eight external ligand sets suggests that MolFilterGAN is useful in triaging or enriching bioactive compounds across a wide range of target types. These results highlighted the importance of MolFilterGAN in evaluating molecules integrally and further accelerating molecular discovery especially combined with advanced AI generative models.

Renal interstitial cells promote nephron regeneration by secreting prostaglandin E2.

In organ regeneration, progenitor and stem cells reside in their native microenvironment, which provides dynamic physical and chemical cues essential to their survival, proliferation, and differentiation. However, the types of cells that form the native microenvironment for renal progenitor cells (RPCs) have not been clarified. Here, single-cell sequencing of zebrafish kidney reveals fabp10a as a principal marker of renal interstitial cells (RICs), which can be specifically labeled by GFP under the control of fabp10a promoter in the fabp10a:GFP transgenic zebrafish. During nephron regeneration, the formation of nephrons is supported by RICs that form a network to wrap the RPC aggregates. RICs that are in close contact with RPC aggregates express cyclooxygenase 2 (Cox2) and secrete prostaglandin E2 (PGE2). Inhibiting PGE2 production prevents nephrogenesis by reducing the proliferation of RPCs. PGE2 cooperates with Wnt4a to promote nephron maturation by regulating ß-catenin stability of RPC aggregates. Overall, these findings indicate that RICs provide a necessary microenvironment for rapid nephrogenesis during nephron regeneration.

Evaluation of sensory and safety quality characteristics of "high mountain tea".

High mountain tea (HT) is widely acknowledged as an essential resource of high-quality tea due to its adaptation to superior ecological environments. In this study, the sensory (aroma and taste) and safety (heavy metals and pesticide residues) characteristics of HT were characterized through sensory evaluation, gas chromatography-mass spectrometry (GC-MS), liquid chromatography-mass spectrometry (LC-MS), flavor activity value, and risk factor analysis. The results elucidated that the aroma sensory characteristics of HT were tender and green, accompanied by sweet and slight chestnut. A total of 8 aroma compounds were identified as the primary substances contributing to the unique aroma characteristics; the difference in the ratio of "green substances" and "chestnut substances" might be the reason for different aroma characteristics in HT and LT (low mountain tea). The taste sensory characteristics of HT were high in freshness and sweetness but low in bitterness and astringency. The high content of soluble sugar (SS), nonester catechins, sweet free amino acids, and low content of caffeine and tea polyphenols were the primary reasons for its taste characteristics. Low temperature stress might be the most fundamental reason for flavor characteristics formation in HT. Furthermore, the pollution risks of 5 heavy metals and 50 pesticide residues in HT were less than 1. The complex ecosystem and low chemical control level were speculated to be the primary reasons for the high safety quality of HT. Overall, these findings provide a more comprehensive understanding of quality characteristics and their formation mechanisms in HT.

MicroRNA miR-627-5p restrains pulmonary artery smooth muscle cell dysfunction by targeting MAP 2 K4 and PI3K/AKT signaling.

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is characterized by pulmonary vascular remodeling, which can be caused by abnormal proliferation and migration of pulmonary artery smooth muscle cells (PASMCs). Several microRNAs were demonstrated to regulate the PASMC dysfunction. Our study intends to evaluate whether miR-627-5p affects cigarette smoke extract (CSE)-induced aberrant biological behaviors of PASMCs. METHODS: PASMCs was treated with CSE to create the in vitro cellular model of COPD. The viability and LDH release of PASMCs was detected by CCK-8 assay and LDH release assay. MiR-627-5p and MAP 2 K4 expression in CSE (2%)-treated PASMCs was detected by qRT-PCR. PASMC proliferation was observed under a microscope, and PASMC migration was assessed by Transwell migration assays. The binding of miR-627-5p on MAP 2 K4 was verified by dual-luciferase reporter assay. Protein levels of MAP2K4 and the PI3K/AKT signaling markers were examined by western blotting. RESULTS: The viability of PASMCs treated with 2% CSE reached a peak. CSE dose-dependently downregulated miR-627-5p expression in PASMCs. MiR-627-5p overexpression attenuated the CSE-induced abnormal proliferation and migration of PASMCs. However, MAP2K4 overexpression antagonized the effects of miR-627-5p on PASMC dysfunction. Importantly, miR-627-5p inhibited CSE-stimulated activation of the PI3K/AKT pathway via downregulating MAP2K4. CONCLUSION: MiR-627-5p improves CSE-induced abnormal proliferation and migration of PASMCs by inhibiting MAP2K4 expression and the PI3K/AKT pathway.

Blood-brain barrier penetration prediction enhanced by uncertainty estimation.

Blood-brain barrier is a pivotal factor to be considered in the process of central nervous system (CNS) drug development, and it is of great significance to rapidly explore the blood-brain barrier permeability (BBBp) of compounds in silico in early drug discovery process. Here, we focus on whether and how uncertainty estimation methods improve in silico BBBp models. We briefly surveyed the current state of in silico BBBp prediction and uncertainty estimation methods of deep learning models, and curated an independent dataset to determine the reliability of the state-of-the-art algorithms. The results exhibit that, despite the comparable performance on BBBp prediction between graph neural networks-based deep learning models and conventional physicochemical-based machine learning models, the GROVER-BBBp model shows greatly improvement when using uncertainty estimations. In particular, the strategy combined Entropy and MC-dropout can increase the accuracy of distinguishing BBB + from BBB - to above 99% by extracting predictions with high confidence level (uncertainty score < 0.1). Case studies on preclinical/clinical drugs for Alzheimer' s disease and marketed antitumor drugs that verified by literature proved the application value of uncertainty estimation enhanced BBBp prediction model, that may facilitate the drug discovery in the field of CNS diseases and metastatic brain tumors.

An inductive graph neural network model for compound-protein interaction prediction based on a homogeneous graph.

Identifying the potential compound-protein interactions (CPIs) plays an essential role in drug development. The computational approaches for CPI prediction can reduce time and costs of experimental methods and have benefited from the continuously improved graph representation learning. However, most of the network-based methods use heterogeneous graphs, which is challenging due to their complex structures and heterogeneous attributes. Therefore, in this work, we transformed the compound-protein heterogeneous graph to a homogeneous graph by integrating the ligand-based protein representations and overall similarity associations. We then proposed an Inductive Graph AggrEgator-based framework, named CPI-IGAE, for CPI prediction. CPI-IGAE learns the low-dimensional representations of compounds and proteins from the homogeneous graph in an end-to-end manner. The results show that CPI-IGAE performs better than some state-of-the-art methods. Further ablation study and visualization of embeddings reveal the advantages of the model architecture and its role in feature extraction, and some of the top ranked CPIs by CPI-IGAE have been validated by a review of recent literature. The data and source codes are available at https://github.com/wanxiaozhe/CPI-IGAE.

Isoorientin Affects Markers of Alzheimer's Disease via Effects on the Oral and Gut Microbiota in APP/PS1 Mice.

BACKGROUND: There is growing evidence of strong associations between the pathogenesis of Alzheimer's disease (AD) and dysbiotic oral and gut microbiota. Recent studies demonstrated that isoorientin (ISO) is anti-inflammatory and alleviates markers of AD, which were hypothesized to be mediated by the oral and gut microbiota. OBJECTIVES: We studied the effects of oral administration of ISO on AD-related markers and the oral and gut microbiota in mice. METHODS: Eight-month-old amyloid precursor protein/presenilin-1 (AP) transgenic male mice were randomly allocated to 3 groups of 15 mice each: vehicle (AP) alone or with a low dose of ISO (AP + ISO-L; 25 mg/kg) or a high dose of ISO (AP + ISO-H; 50 mg/kg). Age-matched wild-type (WT) C57BL/6 male littermates were used as controls. The 4 groups were treated intragastrically with ISO or sterilized ultrapure water for 2 months. AD-related markers in the brain, serum, colon, and liver were analyzed with immunohistochemical and histochemical staining, Western blotting, and ELISA. Oral and gut microbiotas were analyzed using 16S ribosomal RNA gene sequencing. RESULTS: The high-dose ISO treatment significantly decreased amyloid beta 42-positive deposition by 38.1% and 45.2% in the cortex and hippocampus, respectively, of AP mice (P < 0.05). Compared with the AP group, both ISO treatments reduced brain phospho-Tau, phosphor-p65, phosphor-inhibitor of NF-κB, and brain and serum LPS and TNF-α by 17.9%-72.5% and increased brain and serum IL-4 and IL-10 by 130%-210% in the AP + ISO-L and AP + ISO-H groups (P < 0.05). Abundances of 26, 25, and 23 microbial taxa in oral, fecal and cecal samples, respectively, were increased in both the AP + ISO-L and AP + ISO-H groups relative to the AP group [linear discriminant analysis (LDA) >3.0; P < 0.05]. Gram-negative bacteria, Alteromonas, Campylobacterales, and uncultured Bacteroidales bacterium were positively correlated (rho = 0.28-0.59; P < 0.05) with the LPS levels and responses of inflammatory cytokines. CONCLUSIONS: The microbiota-gut-brain axis is a potential mechanism by which ISO reduces AD-related markers in AP mice.

Discovery of Pyrazolo[3,4-d]pyridazinone Derivatives as Selective DDR1 Inhibitors via Deep Learning Based Design, Synthesis, and Biological Evaluation.

Alterations of discoidin domain receptor1 (DDR1) may lead to increased production of inflammatory cytokines, making DDR1 an attractive target for inflammatory bowel disease (IBD) therapy. A scaffold-based molecular design workflow was established and performed by integrating a deep generative model, kinase selectivity screening and molecular docking, leading to a novel DDR1 inhibitor compound 2, which showed potent DDR1 inhibition profile (IC50 = 10.6 ± 1.9 nM) and excellent selectivity against a panel of 430 kinases (S (10) = 0.002 at 0.1 µM). Compound 2 potently inhibited the expression of pro-inflammatory cytokines and DDR1 autophosphorylation in cells, and it also demonstrated promising oral therapeutic effect in a dextran sulfate sodium (DSS)-induced mouse colitis model.

Geniposide protects pulmonary arterial smooth muscle cells from lipopolysaccharide-induced injury via α7nAchR-mediated TLR-4/MyD88 signaling.

Geniposide is a bioactive iridoid glucoside derived from Gardenia jasminoides that has proven anti-inflammatory effects against acute lung injury. The aim of this study was to determine whether geniposide could protect pulmonary arterial smooth muscle cells (PASMCs) from lipopolysaccharide (LPS)-induced injury and to explore the participation of α7 nicotinic acetylcholine receptor (α7nAChR), which was previously reported to suppress pro-inflammatory cytokine production in LPS-stimulated macrophages. In the present study, rat PASMCs were isolated and stimulated using LPS. The effect of geniposide on LPS-induced PASMC injury was then explored. Geniposide exerted anti-apoptotic and anti-inflammatory effects on LPS-treated PASMCs, as demonstrated by the downregulation of pro-apoptotic proteins and pro-inflammatory cytokines, respectively. Furthermore, the α7nAChR agonist PNU282987 accentuated the protective effect of geniposide against LPS-induced injury in PASMCs by inhibiting toll-like receptor-4/myeloid differentiation primary response 88 (TLR-4/MyD88) signaling and downregulating nuclear factor (NF)-κB expression. Conversely, methyllycaconitine, an inhibitor of α7nAChR, attenuated the effects of geniposide. These findings collectively suggested that in conjunction with geniposide, the activation of α7nAChR may contribute to further mitigating LPS-induced PASMC apoptosis and inflammation. In addition, the underlying mechanisms critically involve the NF-κB/MyD88 signaling axis. These results may provide novel insights into the treatment and management of lung diseases via geniposide administration.

Generative Models for De Novo Drug Design.

Artificial intelligence (AI) is booming. Among various AI approaches, generative models have received much attention in recent years. Inspired by these successes, researchers are now applying generative model techniques to de novo drug design, which has been considered as the "holy grail" of drug discovery. In this Perspective, we first focus on describing models such as recurrent neural network, autoencoder, generative adversarial network, transformer, and hybrid models with reinforcement learning. Next, we summarize the applications of generative models to drug design, including generating various compounds to expand the compound library and designing compounds with specific properties, and we also list a few publicly available molecular design tools based on generative models which can be used directly to generate molecules. In addition, we also introduce current benchmarks and metrics frequently used for generative models. Finally, we discuss the challenges and prospects of using generative models to aid drug design.

Isoorientin, a GSK-3ß inhibitor, rescues synaptic dysfunction, spatial memory deficits and attenuates pathological progression in APP/PS1 model mice.

ß-Amyloid (Aß) elevation, tau hyperphosphorylation, and neuroinflammation are major hallmarks of Alzheimer's disease (AD). Glycogen synthase kinase-3ß (GSK-3ß) is a key protein kinase implicated in the pathogenesis of AD. Blockade of GSK-3ß is an attractive therapeutic strategy for AD. Isoorientin, a 6-C-glycosylflavone, was previously shown to be a highly selective inhibitor of GSK-3ß, while exerting neuroprotective effects in neuronal models of AD. In the present study, we evaluated the in vivo effects of isoorientin on GSK-3ß, tau phosphorylation, Aß deposition, neuroinflammatory response, long-term potentiation, and spatial memory in amyloid precursor protein/presenilin 1 (APP/PS1) transgenic mice using biochemical, electrophysiological, and behavioral tests. Chronic oral administration of isoorientin to APP/PS1 mice at 8 months of age attenuated multiple AD pathogenic hallmarks in the brains, including GSK-3ß overactivation, tau hyperphosphorylation, Aß deposition, and neuroinflammation. For neuroinflammation, isoorientin treatment reduced the number of activated microglia associated with Aß-positive plaques, and in parallel reduced the levels of pro-inflammatory factors in the brains of APP/PS1 mice. Strikingly, isoorientin reversed deficits in synaptic long-term potentiation and spatial memory relevant to cognitive functions. Together, the findings suggest that isoorientin is a brain neuroprotector and may be a promising drug lead for treatment of AD and related neurodegenerative disorders.

Inhibitory effect of sinomenine on lung cancer cells via negative regulation of α7 nicotinic acetylcholine receptor.

Lung cancer is the leading cause of cancer deaths worldwide, with a high morbidity and less than 20% survival rate. Therefore, new treatment strategies and drugs are needed to reduce the mortality of patients with lung cancer. α7 nicotinic acetylcholine receptor (α7 nAChR), as a receptor of nicotine and its metabolites, is a potential target for lung cancer treatment. Our previous studies revealed that sinomenine plays anti-inflammation roles via α7 nAChR and down-regulates the expression of this receptor, thus increasing the inflammatory response. Hence, sinomenine is possibly a natural ligand of this receptor. In the present study, the effects of sinomenine on lung cancer A549 cells and tumor-bearing mice were determined to investigate whether this alkaloid has an inhibitory effect on lung cancer via α7 nAChR. CCK-8 assay, wound-healing test, and flow cytometry were performed for cell proliferation, cell migration, and apoptosis analysis in vitro, respectively. Xenograft mice were used to evaluate the effects of sinomenine in vivo. Results showed that sinomenine decreased cell proliferation and migration abilities but increased the percentage of apoptotic cells. Tumor volume in tumor-bearing mice was significantly reduced after sinomenine treatment compared with that in the vehicle group mice (p < 0.05). Furthermore, the effects of sinomenine were abolished by the α7 nAChR antagonist mecamylamine and the allosteric modulator PNU-120596, but no change occurred when the mice were pretreated with the muscarinic acetylcholine receptor antagonist atropine. Meanwhile, sinomenine suppressed α7 nAChR expression in vitro and in vivo, as well as the related signaling molecules pERK1/2 and ERK1/2 and the transcription factors TTF-1 and SP-1. By contrast, sinomenine up-regulated the expression of another transcription factor, Egr-1. These effects were restricted by mecamylamine and PNU but not by atropine. Results suggested that sinomenine can inhibit lung cancer via α7 nAChR in a negative feedback mode.

DrugSpaceX: a large screenable and synthetically tractable database extending drug space.

One of the most prominent topics in drug discovery is efficient exploration of the vast drug-like chemical space to find synthesizable and novel chemical structures with desired biological properties. To address this challenge, we created the DrugSpaceX (https://drugspacex.simm.ac.cn/) database based on expert-defined transformations of approved drug molecules. The current version of DrugSpaceX contains >100 million transformed chemical products for virtual screening, with outstanding characteristics in terms of structural novelty, diversity and large three-dimensional chemical space coverage. To illustrate its practical application in drug discovery, we used a case study of discoidin domain receptor 1 (DDR1), a kinase target implicated in fibrosis and other diseases, to show DrugSpaceX performing a quick search of initial hit compounds. Additionally, for ligand identification and optimization purposes, DrugSpaceX also provides several subsets for download, including a 10% diversity subset, an extended drug-like subset, a drug-like subset, a lead-like subset, and a fragment-like subset. In addition to chemical properties and transformation instructions, DrugSpaceX can locate the position of transformation, which will enable medicinal chemists to easily integrate strategy planning and protection design.

Isoorientin Inhibits Inflammation in Macrophages and Endotoxemia Mice by Regulating Glycogen Synthase Kinase 3ß.

Isoorientin has anti-inflammatory effects; however, the mechanism remains unclear. We previously found isoorientin is an inhibitor of glycogen synthase kinase 3ß (GSK3ß) in vitro. Overactivation of GSK3ß is associated with inflammatory responses. GSK3ß is inactivated by phosphorylation at Ser9 (i.e., p-GSK3ß). Lithium chloride (LiCl) inhibits GSK3ß and also increases p-GSK3ß (Ser9). The present study investigated the anti-inflammatory effect and mechanism of isoorientin via GSK3ß regulation in lipopolysaccharide- (LPS-) induced RAW264.7 murine macrophage-like cells and endotoxemia mice. LiCl was used as a control. While AKT phosphorylates GSK3ß, MK-2206, a selective AKT inhibitor, was used to activate GSK3ß via AKT inhibition (i.e., not phosphorylate GSK3ß at Ser9). The proinflammatory cytokines TNF-α, IL-6, and IL-1ß were detected by ELISA or quantitative real-time PCR, while COX-2 by Western blotting. The p-GSK3ß and GSK3ß downstream signal molecules, including NF-κB, ERK, Nrf2, and HO-1, as well as the tight junction proteins ZO-1 and occludin were measured by Western blotting. The results showed that isoorientin decreased the production of TNF-α, IL-6, and IL-1ß and increased the expression of p-GSK3ß in vitro and in vivo, similar to LiCl. Coadministration of isoorientin and LiCl showed antagonistic effects. Isoorientin decreased the expression of COX-2, inhibited the activation of ERK and NF-κB, and increased the activation of Nrf2/HO-1 in LPS-induced RAW264.7 cells. Isoorientin increased the expressions of occludin and ZO-1 in the brain of endotoxemia mice. In summary, isoorientin can inhibit GSK3ß by increasing p-GSK3ß and regulate the downstream signal molecules to inhibit inflammation and protect the integrity of the blood-brain barrier and the homeostasis in the brain.

Automated design and optimization of multitarget schizophrenia drug candidates by deep learning.

Complex neuropsychiatric diseases such as schizophrenia require drugs that can target multiple G protein-coupled receptors (GPCRs) to modulate complex neuropsychiatric functions. Here, we report an automated system comprising a deep recurrent neural network (RNN) and a multitask deep neural network (MTDNN) to design and optimize multitarget antipsychotic drugs. The system has successfully generated novel molecule structures with desired multiple target activities, among which high-ranking compound 3 was synthesized, and demonstrated potent activities against dopamine D2, serotonin 5-HT1A and 5-HT2A receptors. Hit expansion based on the MTDNN was performed, 6 analogs of compound 3 were evaluated experimentally, among which compound 8 not only exhibited specific polypharmacology profiles but also showed antipsychotic effect in animal models with low potential for sedation and catalepsy, highlighting their suitability for further preclinical studies. The approach can be an efficient tool for designing lead compounds with multitarget profiles to achieve the desired efficacy in the treatment of complex neuropsychiatric diseases.

TransformerCPI: improving compound-protein interaction prediction by sequence-based deep learning with self-attention mechanism and label reversal experiments.

MOTIVATION: Identifying compound-protein interaction (CPI) is a crucial task in drug discovery and chemogenomics studies, and proteins without three-dimensional structure account for a large part of potential biological targets, which requires developing methods using only protein sequence information to predict CPI. However, sequence-based CPI models may face some specific pitfalls, including using inappropriate datasets, hidden ligand bias and splitting datasets inappropriately, resulting in overestimation of their prediction performance. RESULTS: To address these issues, we here constructed new datasets specific for CPI prediction, proposed a novel transformer neural network named TransformerCPI, and introduced a more rigorous label reversal experiment to test whether a model learns true interaction features. TransformerCPI achieved much improved performance on the new experiments, and it can be deconvolved to highlight important interacting regions of protein sequences and compound atoms, which may contribute chemical biology studies with useful guidance for further ligand structural optimization. AVAILABILITY AND IMPLEMENTATION: https://github.com/lifanchen-simm/transformerCPI.

Ethylcellulose-based drug nano depots fabricated using a modified triaxial electrospinning.

New strategies based on advanced technologies are highly desired for expanding the applications of biological macromolecules in the applied scientific fields. In the present study, a new kind of core-shell nano depots were designed, in which the shell section was a drug-polymer composite and the core section was a drug reservoir. With ethyl cellulose and ketoprofen as a filament-forming polymeric matrix and a model drug, respectively, a triaxial electrospinning apparatus was developed to conduct both coaxial and triaxial processes, by which monolithic nanofibers F1 and core-shell nano depots F2 were successfully prepared. Although both of them had the same double components, their different nanostructures generated considerable differences in providing drug sustained release profiles. The core-shell nanofiber depots F2 were able to provide a better zero-order drug release profile: no initial burst release, smooth sustained release effect, and smaller tailing-off release for a nice zero-order drug release kinetics. The release percentage (Q) can be linearly manipulated through the release time (t) according to the equation Q2 = 9.40 + 4.74 t (R = 0.9936), providing opportunity for precise administration. The developed strategy and advanced electrospinning technique exhibit a new way for constructing process-structure-performance relationships at nano scale and for expanding the potential applications of biological macromolecules in the applied fields.