GNNGL-PPI: multi-category prediction of protein-protein interactions using graph neural networks based on global graphs and local subgraphs.

Most proteins exert their functions by interacting with other proteins, making the identification of protein-protein interactions (PPI) crucial for understanding biological activities, pathological mechanisms, and clinical therapies. Developing effective and reliable computational methods for predicting PPI can significantly reduce the time-consuming and labor-intensive associated traditional biological experiments. However, accurately identifying the specific categories of protein-protein interactions and improving the prediction accuracy of the computational methods remain dual challenges. To tackle these challenges, we proposed a novel graph neural network method called GNNGL-PPI for multi-category prediction of PPI based on global graphs and local subgraphs. GNNGL-PPI consisted of two main components: using Graph Isomorphism Network (GIN) to extract global graph features from PPI network graph, and employing GIN As Kernel (GIN-AK) to extract local subgraph features from the subgraphs of protein vertices. Additionally, considering the imbalanced distribution of samples in each category within the benchmark datasets, we introduced an Asymmetric Loss (ASL) function to further enhance the predictive performance of the method. Through evaluations on six benchmark test sets formed by three different dataset partitioning algorithms (Random, BFS, DFS), GNNGL-PPI outperformed the state-of-the-art multi-category prediction methods of PPI, as measured by the comprehensive performance evaluation metric F1-measure. Furthermore, interpretability analysis confirmed the effectiveness of GNNGL-PPI as a reliable multi-category prediction method for predicting protein-protein interactions.

A comprehensive review of the recent advances on predicting drug-target affinity based on deep learning.

Accurate calculation of drug-target affinity (DTA) is crucial for various applications in the pharmaceutical industry, including drug screening, design, and repurposing. However, traditional machine learning methods for calculating DTA often lack accuracy, posing a significant challenge in accurately predicting DTA. Fortunately, deep learning has emerged as a promising approach in computational biology, leading to the development of various deep learning-based methods for DTA prediction. To support researchers in developing novel and highly precision methods, we have provided a comprehensive review of recent advances in predicting DTA using deep learning. We firstly conducted a statistical analysis of commonly used public datasets, providing essential information and introducing the used fields of these datasets. We further explored the common representations of sequences and structures of drugs and targets. These analyses served as the foundation for constructing DTA prediction methods based on deep learning. Next, we focused on explaining how deep learning models, such as Convolutional Neural Networks (CNNs), Recurrent Neural Networks (RNNs), Transformer, and Graph Neural Networks (GNNs), were effectively employed in specific DTA prediction methods. We highlighted the unique advantages and applications of these models in the context of DTA prediction. Finally, we conducted a performance analysis of multiple state-of-the-art methods for predicting DTA based on deep learning. The comprehensive review aimed to help researchers understand the shortcomings and advantages of existing methods, and further develop high-precision DTA prediction tool to promote the development of drug discovery.

Drug-Online: an online platform for drug-target interaction, affinity, and binding sites identification using deep learning.

BACKGROUND: Accurately identifying drug-target interaction (DTI), affinity (DTA), and binding sites (DTS) is crucial for drug screening, repositioning, and design, as well as for understanding the functions of target. Although there are a few online platforms based on deep learning for drug-target interaction, affinity, and binding sites identification, there is currently no integrated online platforms for all three aspects. RESULTS: Our solution, the novel integrated online platform Drug-Online, has been developed to facilitate drug screening, target identification, and understanding the functions of target in a progressive manner of "interaction-affinity-binding sites". Drug-Online platform consists of three parts: the first part uses the drug-target interaction identification method MGraphDTA, based on graph neural networks (GNN) and convolutional neural networks (CNN), to identify whether there is a drug-target interaction. If an interaction is identified, the second part employs the drug-target affinity identification method MMDTA, also based on GNN and CNN, to calculate the strength of drug-target interaction, i.e., affinity. Finally, the third part identifies drug-target binding sites, i.e., pockets. The method pt-lm-gnn used in this part is also based on GNN. CONCLUSIONS: Drug-Online is a reliable online platform that integrates drug-target interaction, affinity, and binding sites identification. It is freely available via the Internet at http://39.106.7.26:8000/Drug-Online/ .

MMDTA: A Multimodal Deep Model for Drug-Target Affinity with a Hybrid Fusion Strategy.

The prediction of the drug-target affinity (DTA) plays an important role in evaluating molecular druggability. Although deep learning-based models for DTA prediction have been extensively attempted, there are rare reports on multimodal models that leverage various fusion strategies to exploit heterogeneous information from multiple different modalities of drugs and targets. In this study, we proposed a multimodal deep model named MMDTA, which integrated the heterogeneous information from various modalities of drugs and targets using a hybrid fusion strategy to enhance DTA prediction. To achieve this, MMDTA first employed convolutional neural networks (CNNs) and graph convolutional networks (GCNs) to extract diverse heterogeneous information from the sequences and structures of drugs and targets. It then utilized a hybrid fusion strategy to combine and complement the extracted heterogeneous information, resulting in the fused modal information for predicting drug-target affinity through the fully connected (FC) layers. Experimental results demonstrated that MMDTA outperformed the competitive state-of-the-art deep learning models on the widely used benchmark data sets, particularly with a significantly improved key evaluation metric, Root Mean Square Error (RMSE). Furthermore, MMDTA exhibited excellent generalization and practical application performance on multiple different data sets. These findings highlighted MMDTA's accuracy and reliability in predicting the drug-target binding affinity. For researchers interested in the source data and code, they are accessible at http://github.com/dldxzx/MMDTA.

Viridibacillus soli sp. nov., isolated from forest soil in Ailaoshan National Nature Reserve.

A Gram stain-positive, rod-shaped, and subterminal endospore-forming bacterium, designated strain YIM B01967T, was isolated from a forest soil sample collected in Ailaoshan National Nature Reserve, Yuxi City, Xinpin county, Yunnan province, China. Strain YIM B01967T showed the highest 16S rRNA gene sequence similarity with Viridibacillus arvi (99.1%) and Viridibacillus arenosi (98.9%). Based on the phylogenetic and 16S rRNA gene sequence results, strain YIM B01967T was affiliated to the genus Viridibacillus. The growth of YIM B01967T was observed at 15-35 °C (optimum, 28 °C), pH 7.0-9.0 (optimum, pH 7.5) and in the presence of 0-2% (w/v) NaCl (optimum in 2% NaCl). The cell wall sugars include ribose, glucose, arabinose, galactose, and mannose. The quinone system consisted of the major compound MK-8 and moderate amounts of MK-7. The major fatty acids (> 10%) included iso-C15:0, anteiso-C15:0, C16:1 ω10c. The major polar lipids profile included DPG, PME. The cell wall peptidoglycan was most likely of the type A4α with an L-Lys-D-Asp interpeptide bridge. The genomic DNA G + C content of strain YIM B01967T was 36.3 mol%. The ANI and digital DNA-DNA hybridization (dDDH) values between strain YIM B01967T and Viridibacillus arvi DSM 16317 T, Viridibacillus arenosi DSM 16319 T were 61.0% and 32.1%, 60.0% and 33.1% based on the draft genome sequence. The results support the conclusion that strain YIM B01967T represents a novel species of the genus Viridibacillus, for which the name Viridibacillus soli sp. nov., is proposed. The type strain is YIM B01967T (= KCTC 43249 T = CGMCC 1.18436 T).

Propioniciclava soli sp. nov., isolated from forest soil, Yunnan, China, and reclassification of the genus Brevilactibacter into the genus Propioniciclava, and Brevilactibacter sinopodophylli, Brevilactibacter flavus, and Brevilactibacter coleopterorum as Propioniciclava sinopodophylli comb. nov., Propioniciclava flava comb. nov., and Propioniciclava coleopterorum comb. nov., respectively.

A Gram-stain-positive, coccus-shaped, facultatively anaerobic, non-motile bacterial strain, designated YIM S02567T, was isolated from a forest soil sample collected from Gejiu City, Yunnan Province, southwest PR China. Growth was observed at 10-45 °C, at pH 6.0-9.5, in the presence of up to 4.0% (w/v) NaCl on R2A medium. The results of 16S rRNA gene sequence similarity analysis showed that strain YIM S02567T was most closely related to the type strain of Brevilactibacter sinopodophylli (95.4%) and Propioniciclava tarda (94.7%), and phylogenetic analysis based on genome data showed that strain YIM S02567T should be assigned to the genus Propioniciclava. The cell-wall diamino acid was meso-diaminopimelic acid. The major cellular fatty acids were identified as anteiso-C15:0 and C16:0, and the major polar lipids were diphosphatidylglycerol, phosphatidylglycerol, and two unidentified glycolipids. The predominant menaquinone was MK-9(H4). The genomic DNA G + C content was 71.2 mol%. Based on the polyphasic taxonomic evidence, strain YIM S02567T is assigned to a novel member of the genus Propioniciclava, for which the name Propioniciclava soli sp. nov., (type strain YIM S02567T = CCTCC AB 2020128T = CGMCC 1.18504T = KCTC 49478T) is proposed. Furthermore, we propose the reclassification of Brevilactibacter as Propioniciclava gen. nov.

Functional Characterization of a New Bifunctional Terpene Synthase LpNES1 from a Medicinal Plant Laggera pter odonta.

Laggera pterodonta, known in China as 'Choulingdan' for its stimulous odor, has long been used as traditional herbal medicine. The essential oil of L. pterodonta, which exhibits various pharmacological activities, is a rich resource of monoterpenes and sesquiterpenes. To date, however, the terpene synthases responsible for their production remain unknown. In present study, a new terpene synthase gene (LpNES1) was identified from L. pterodonta, transcript level of which was significantly upregulated in response to methyl jasmonate treatment. Recombinant LpNES1 could synthesize (E)-nerolidol and minor ß-farnesene from farnesyl diphosphate and linalool from geranyl diphosphate in vitro. Whereas, only sesquiterpenes including (E)-nerolidol and minor ß-farnesene were released when LpNES1 was reconstituted in yeast, even coexpressed with a geranyl diphosphate synthase (ERG20WW). Combined with subcellular localization experiment, the result indicated that the cytosol-targeted LpNES1 was responsible for (E)-nerolidol biosynthesis exclusively in L. pterodonta. Additionally, the expression level of LpNES1 gene was more prominent in floral buds than that in other tissues. LpNES1 characterized in present study not only lays the molecular foundation for sesquiterpene biosynthesis of L. pterodonta, but provides a key element for further biosynthesis of bioactive compound in microbes.

Aestuariimicrobium ganziense sp. nov., a new Gram-positive bacterium isolated from soil in the Ganzi Tibetan autonomous prefecture, China.

A novel Gram-stain positive, oval-shaped, and non-flagellated bacterium, designated YIM S02566T, was isolated from alpine soil in Shadui Towns, Ganzi County, Ganzi Tibetan Autonomous Prefecture, Sichuan Province, PR China. Growth occurred at 23-35 °C (optimum, 30 °C) in the presence of 0.5-4% (w/v) NaCl (optimum, 1%) and at pH 7.0-8.0 (optimum, pH 7.0). The phylogenetic analysis based on 16S rRNA gene sequence revealed that strain YIM S02566T was most closely related to the genus Aestuariimicrobium, with Aestuariimicrobium kwangyangense R27T and Aestuariimicrobium soli D6T as its closest relative (sequence similarities were 96.3% and 95.4%, respectively). YIM S02566T contained LL-diaminopimelic acid in the cell wall. MK-9(H4) was the predominant menaquinone. The major fatty acid patterns were anteiso-C15:0 (60.0%). The major polar lipid was DPG. The genome size of strain YIM S02566T was 3.1 Mb, comprising 3078 predicted genes with a DNA G + C content of 69.0 mol%. Based on these genotypic, chemotaxonomic and phenotypic evidences, strain YIM S02566T was identified as a novel species in the genus Aestuariimicrobium, for which the name Aestuariimicrobium ganziense sp. nov. is proposed. The type strain is YIM S02566T (= CGMCC 1.18751 T = KCTC 49,477 T).

[Recent advances in biosynthesis of forskolin].

Forskolin is a complex labdane plant diterpenoid, which has been used in the treatment of a variety of diseases based on its activity as an activator of adenosine monophosphate(cAMP) cyclase. Natural forskolin exists only in the cork layer of the root of Coleus forskohlii. Due to the complexity of the extraction and chemical synthesis processes, the yield and purity of forskolin cannot meet commercial requirements. In recent years, with the rapid development of synthetic biology and the analysis and interpretation of many diterpene biosynthetic pathways, a new approach has been provided for the green production of forskolin. In this paper, the structure, activity, biosynthetic pathway and the heterologous biosynthesis of forskolin were reviewed. The problems and solutions in the heterologous biosynthesis of forskolin were also discussed and summarized, which will provide references for the construction of high-yielding forskolin engineering strains.

Cloning and Characterization of a Ginsenoside-Hydrolyzing α-L-Arabinofuranosidase, CaAraf51, From Cellulosimicrobium aquatile Lyp51.

Moutai Jiuqu is a famous aromatic raw material of Maotai flavor liquor in China. It is brewed at high temperature and contains many kinds of bacteria, molds, and yeasts. There are many useful glycoside hydrolases in these microfloras, from which efficient glycoside hydrolases can be screened for biotransformation of natural saponins. In this study, an α-L-arabinofuranosidase gene (CaAraf51, 1524 bp, 507 amino acid, 55.07 kDa, and pI = 4.8) was cloned from Cellulosimicrobium aquatile Lyp51, which was isolated from the Maotai Jiuqu. The CaAraf51 was heterogeneously expressed in E. coli BL21 (DE3) and purified by N-terminal His-tag with the Ni2+-affinity column chromatography. The results show that purified CaAraf51 has a 6.8-fold purification factor and specific activity of 15 U/mg. Under optimal conditions (pH 5.0, temperature 40 °C), kinetic parameters Km of CaAraf51 for pNPαAraf and Rc were 1.1 and 0.57 mM, the Vmax were 25 and 6.25 µmol/min/mg, respectively. 90% of 0.87 mg Rc substrate can be transformed by 9.6 U purified CaAraf51 in 1 mL reaction system under suitable conditions (30 °C, pH 7.5 phosphate buffer, 1 h). In addition, we also tested the effects of metal ions and chemical agents on the activity of CaAraf51. According to systematically studied its function and enzymatic properties, CaAraf51 has excellent value and potential of biotransformation Rc into Rd.

Repeated evolution of cytochrome P450-mediated spiroketal steroid biosynthesis in plants.

Diosgenin is a spiroketal steroidal natural product extracted from plants and used as the single most important precursor for the world steroid hormone industry. The sporadic occurrences of diosgenin in distantly related plants imply possible independent biosynthetic origins. The characteristic 5,6-spiroketal moiety in diosgenin is reminiscent of the spiroketal moiety present in anthelmintic avermectins isolated from actinomycete bacteria. How plants gained the ability to biosynthesize spiroketal natural products is unknown. Here, we report the diosgenin-biosynthetic pathways in himalayan paris (Paris polyphylla), a monocot medicinal plant with hemostatic and antibacterial properties, and fenugreek (Trigonella foenum-graecum), an eudicot culinary herb plant commonly used as a galactagogue. Both plants have independently recruited pairs of cytochromes P450 that catalyze oxidative 5,6-spiroketalization of cholesterol to produce diosgenin, with evolutionary progenitors traced to conserved phytohormone metabolism. This study paves the way for engineering the production of diosgenin and derived analogs in heterologous hosts.

A new pregnane steroid from cultures of Aspergillus versicolor.

A new pregnane steroid, named aspergillon A (1), together with two known compounds, (22E,24R)-ergosta-5,7,22-trien-3ß-ol (2) and (22E, 24R)-ergosta-4,6,8(14),22-tetraen-3-one (3) were isolated from cultures of the tin mine tailings-associated fungus Aspergillus versicolor. The new structure and absolute configuration were determined with the help of extensive spectroscopic analyses and quantum chemical calculations of the electronic circular dichroism (ECD) spectra.

A new cyclohexenone from the tin mine tailings-derived fungus Aspergillus flavus YIM DT 10012.

A new cyclohexenone, named phomaligol D (1), together with two known compounds, kojic acid (2) and phomaligol A (3) were isolated from the tin mine tailings-derived fungus Aspergillus flavus YIM DT 10012. Their structures were elucidated by detailed analysis of spectroscopic data.

A novel phenylspirodrimane dimer from cultures of the fungus Stachybotrys chartarum.

Stachartone A (1), a novel phenylspirodrimane dimer, was isolated from cultures of the tin mine tailings-associated fungus Stachybotrys chartarum. Its structure was elucidated by means of spectroscopic methods. At the same time, compound (1) was tested for its cytotoxicity against five human cancer cell lines.

A new phenylspirodrimane dimer from the fungus Stachybotrys chartarum.

A new phenylspirodrimane dimer, named stachartarin A (1), was isolated from cultures of the tin mine tailings-associated fungus Stachybotrys chartarum. Its structures were elucidated by means of spectroscopic methods. At the same time, the compound was tested for its cytotoxicity against HL-60, SMMC-7721, A-549, MCF-7, and SW480 cells.

Ameliorative effects of Compound K and ginsenoside Rh1 on non-alcoholic fatty liver disease in rats.

Non-alcoholic fatty liver disease (NAFLD) is a common liver disease, which has no standard treatment available. Panax notoginseng saponines (PNS) have recently been reported to protect liver against hepatocyte injury induced by ethanol or high fat diet (HFD) in rats. Compound K and ginsenoside Rh1 are the main metabolites of PNS. In this study, we evaluated the effects of CK and Rh1 on NAFLD. Rats fed HFD showed significant elevations in liver function markers, lipids, glucose tolerance, and insulin resistance. Treatment with CK or Rh1 either alone or in combination dramatically ameliorated the liver function impairment induced by HFD. Histologically, CK and Rh1 significantly reversed HFD-induced hepatocyte injury and liver fibrosis. In vitro experiments demonstrated that treatment with CK or Rh1 alone or in combination markedly induced cell apoptosis, and inhibited cell proliferation and activation in HSC-T6 cells. Additionally, CK and Rh1, either alone or in combination, also repressed the expression of fibrotic factors TIMP-1, PC-I, and PC-III. Taken together, our results demonstrate that CK and Rh1 have positive effects on NAFLD via the anti-fibrotic and hepatoprotective activity.

Two new diketopiperazines from the tin mine tailings-derived fungus Schizophyllum commune YIM DT 10058.

Two new diketopiperazines, named diphenylalazine C (1) and epicoccin U (2), together with six known compounds (3-8) were isolated from the EtOAc extract of the fermentation broth of the tin mine tailings-derived fungus Schizophyllum commune YIM DT 10058. Their structures were elucidated by detailed analysis of spectroscopic data and comparison with related known compounds. Compounds 1 and 2 exhibited weak antibacterial and cytotoxic activities.

A new quinolinone from freshwater lake-derived fungus Myrothecium verrucaria.

One new quinolinone, 7-hydroxy-3-methoxyviridicatin (1), along with eight known compounds (2-9) was isolated from the fungus Myrothecium verrucaria, which was collected from lake water of Chenghai, Yunnan Province, China. Their structures were elucidated by detailed analysis of spectroscopic data and comparison with related known compounds. Compounds 1 and 2 exhibited weak antibacterial activity. To the best of our knowledge, this is the first report on quinolinones (1-4) as the secondary metabolites of M. verrucaria.

Identification of anti-inflammatory fractions of Geranium wilfordii using tumor necrosis factor-alpha as a drug target on Herbochip® - an array-based high throughput screening platform.

BACKGROUND: Geranium wilfordii is one of the major species used as Herba Geranii (lao-guan-cao) in China, it is commonly used solely or in polyherbal formulations for treatment of joint pain resulted from rheumatoid arthritis (RA) and gout. This herb is used to validate a target-based drug screening platform called Herbochip® and evaluate anti-inflammatory effects of Geranium wilfordii ethanolic extract (GWE) using tumor necrosis factor-alpha (TNF-α) as a drug target together with subsequent in vitro and in vivo assays. METHODS: A microarray-based drug screening platform was constructed by arraying HPLC fractions of herbal extracts onto a surface-activated polystyrene slide (Herbochip®). Using TNF-α as a molecular probe, fractions of 82 selected herbal extracts, including GWE, were then screened to identify plant extracts containing TNF-α-binding agents. Cytotoxicity of GWE and modulatory effects of GWE on TNF-α expression were evaluated by cell-based assays using TNF-α sensitive murine fibrosarcoma L929 cells as an in vitro model. RESULTS: The in vivo anti-inflammatory effects of GWE were further assessed by animal models including carrageenan-induced hind paw edema in rats and xylene-induced ear edema in mice, in comparison with aspirin. The hybridization data obtained by Herbochip® analysis showed unambiguous signals which confirmed TNF-α binding activity in 46 herbal extracts including GWE. In L929 cells GWE showed significant inhibitory effect on TNF-α expression with negligible cytotoxicity. GWE also significantly inhibited formation of carrageenan-induced hind paw edema and xylene-induced ear edema in animal models, indicating that it indeed possessed anti-inflammatory activity. CONCLUSION: We have thus validated effectiveness of the Herbochip® drug screening platform using TNF-α as a molecular target. Subsequent experiments on GWE lead us to conclude that the anti-RA activity of GWE can be attributed to inhibitory effect of GWE on the key inflammatory factor, TNF-α. Our results contribute towards validation of the traditional use of GWE in the treatment of RA and other inflammatory joint disorders.

Gracilimonas mengyeensis sp. nov., a moderately halophilic bacterium isolated from a salt mine in Yunnan, south-western China.

A facultatively anaerobic, Gram-staining-negative, pale red-pigmented, non-motile, rod-shaped, moderately halophilic bacterium, designated strain YIM J14(T), was isolated from a sediment sample from a salt mine in Yunnan, south-western China. Growth occurred at NaCl concentrations of between 2 % and 15 % (w/v) and optimally with 5-9 % NaCl. The optimum temperature and pH for growth of the strain were 28 °C and pH 7.5. The major cellular fatty acids were iso-C15 : 0, anteiso-C15 : 0 and iso-C17 : 1ω9c/10-methyl-C16 : 0. The polar lipid profile was composed predominantly of diphosphatidylglycerol, phosphatidylcholine and one unknown phospholipid. Minor amounts of other lipids were also detectable. The genomic DNA G+C content was 47.2 mol%. Phylogenetic analysis based on 16S rRNA gene sequence comparisons indicated that strain YIM J14(T) was related to Gracilimonas tropica in the phylum Bacteroidetes. The level of 16S rRNA gene sequence similarity between strain YIM J14(T) and Gracilimonas tropica CL-CB462(T) was 96.9 %. A DNA-DNA hybridization experiment between strain YIM J14(T) and Gracilimonas tropica indicated levels of relatedness of 28 %. Chemotaxonomic data supported the placement of strain YIM J14(T) in the genus Gracilimonas. DNA-DNA hybridization and biochemical and physiological characterization allowed strain YIM J14(T) to be differentiated from Gracilimonas tropica. It is therefore considered to represent a novel species of the genus Gracilimonas, for which the name Gracilimonas mengyeensis sp. nov. is proposed. The type strain YIM J14(T) ( = ACCC 10717(T) = DSM 21985(T)).