PhosBERT: A self-supervised learning model for identifying phosphorylation sites in SARS-CoV-2-infected human cells.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is a single-stranded RNA virus, which mainly causes respiratory and enteric diseases and is responsible for the outbreak of coronavirus disease 19 (COVID-19). Numerous studies have demonstrated that SARS-CoV-2 infection will lead to a significant dysregulation of protein post-translational modification profile in human cells. The accurate recognition of phosphorylation sites in host cells will contribute to a deep understanding of the pathogenic mechanisms of SARS-CoV-2 and also help to screen drugs and compounds with antiviral potential. Therefore, there is a need to develop cost-effective and high-precision computational strategies for specifically identifying SARS-CoV-2-infected phosphorylation sites. In this work, we first implemented a custom neural network model (named PhosBERT) on the basis of a pre-trained protein language model of ProtBert, which was a self-supervised learning approach developed on the Bidirectional Encoder Representation from Transformers (BERT) architecture. PhosBERT was then trained and validated on serine (S) and threonine (T) phosphorylation dataset and tyrosine (Y) phosphorylation dataset with 5-fold cross-validation, respectively. Independent validation results showed that PhosBERT could identify S/T phosphorylation sites with high accuracy and AUC (area under the receiver operating characteristic) value of 81.9% and 0.896. The prediction accuracy and AUC value of Y phosphorylation sites reached up to 87.1% and 0.902. It indicated that the proposed model was of good prediction ability and stability and would provide a new approach for studying SARS-CoV-2 phosphorylation sites.

Prediction of blood-brain barrier penetrating peptides based on data augmentation with Augur.

BACKGROUND: The blood-brain barrier serves as a critical interface between the bloodstream and brain tissue, mainly composed of pericytes, neurons, endothelial cells, and tightly connected basal membranes. It plays a pivotal role in safeguarding brain from harmful substances, thus protecting the integrity of the nervous system and preserving overall brain homeostasis. However, this remarkable selective transmission also poses a formidable challenge in the realm of central nervous system diseases treatment, hindering the delivery of large-molecule drugs into the brain. In response to this challenge, many researchers have devoted themselves to developing drug delivery systems capable of breaching the blood-brain barrier. Among these, blood-brain barrier penetrating peptides have emerged as promising candidates. These peptides had the advantages of high biosafety, ease of synthesis, and exceptional penetration efficiency, making them an effective drug delivery solution. While previous studies have developed a few prediction models for blood-brain barrier penetrating peptides, their performance has often been hampered by issue of limited positive data. RESULTS: In this study, we present Augur, a novel prediction model using borderline-SMOTE-based data augmentation and machine learning. we extract highly interpretable physicochemical properties of blood-brain barrier penetrating peptides while solving the issues of small sample size and imbalance of positive and negative samples. Experimental results demonstrate the superior prediction performance of Augur with an AUC value of 0.932 on the training set and 0.931 on the independent test set. CONCLUSIONS: This newly developed Augur model demonstrates superior performance in predicting blood-brain barrier penetrating peptides, offering valuable insights for drug development targeting neurological disorders. This breakthrough may enhance the efficiency of peptide-based drug discovery and pave the way for innovative treatment strategies for central nervous system diseases.

Isolation and Identification of Salinity-Tolerant Rhizobia and Nodulation Phenotype Analysis in Different Soybean Germplasms.

Increasing the soybean-planting area and increasing the soybean yield per unit area are two effective solutions to improve the overall soybean yield. Northeast China has a large saline soil area, and if soybeans could be grown there with the help of isolated saline-tolerant rhizobia, the soybean cultivation area in China could be effectively expanded. In this study, soybeans were planted in soils at different latitudes in China, and four strains of rhizobia were isolated and identified from the soybean nodules. According to the latitudes of the soil-sampling sites from high to low, the four isolated strains were identified as HLNEAU1, HLNEAU2, HLNEAU3, and HLNEAU4. In this study, the isolated strains were identified for their resistances, and their acid and saline tolerances and nitrogen fixation capacities were preliminarily identified. Ten representative soybean germplasm resources in Northeast China were inoculated with these four strains, and the compatibilities of these four rhizobium strains with the soybean germplasm resources were analyzed. All four isolates were able to establish different extents of compatibility with 10 soybean resources. Hefeng 50 had good compatibility with the four isolated strains, while Suinong 14 showed the best compatibility with HLNEAU2. The isolated rhizobacteria could successfully establish symbiosis with the soybeans, but host specificity was also present. This study was a preliminary exploration of the use of salinity-tolerant rhizobacteria to help the soybean nitrogen fixation in saline soils in order to increase the soybean acreage, and it provides a valuable theoretical basis for the application of saline-tolerant rhizobia.

PSnoD: identifying potential snoRNA-disease associations based on bounded nuclear norm regularization.

Many studies have proved that small nucleolar RNAs (snoRNAs) play critical roles in the development of various human complex diseases. Discovering the associations between snoRNAs and diseases is an important step toward understanding the pathogenesis and characteristics of diseases. However, uncovering associations via traditional experimental approaches is costly and time-consuming. This study proposed a bounded nuclear norm regularization-based method, called PSnoD, to predict snoRNA-disease associations. Benchmark experiments showed that compared with the state-of-the-art methods, PSnoD achieved a superior performance in the 5-fold stratified shuffle split. PSnoD produced a robust performance with an area under receiver-operating characteristic of 0.90 and an area under precision-recall of 0.55, highlighting the effectiveness of our proposed method. In addition, the computational efficiency of PSnoD was also demonstrated by comparison with other matrix completion techniques. More importantly, the case study further elucidated the ability of PSnoD to screen potential snoRNA-disease associations. The code of PSnoD has been uploaded to https://github.com/linDing-groups/PSnoD. Based on PSnoD, we established a web server that is freely accessed via http://psnod.lin-group.cn/.

iRice-MS: An integrated XGBoost model for detecting multitype post-translational modification sites in rice.

Post-translational modification (PTM) refers to the covalent and enzymatic modification of proteins after protein biosynthesis, which orchestrates a variety of biological processes. Detecting PTM sites in proteome scale is one of the key steps to in-depth understanding their regulation mechanisms. In this study, we presented an integrated method based on eXtreme Gradient Boosting (XGBoost), called iRice-MS, to identify 2-hydroxyisobutyrylation, crotonylation, malonylation, ubiquitination, succinylation and acetylation in rice. For each PTM-specific model, we adopted eight feature encoding schemes, including sequence-based features, physicochemical property-based features and spatial mapping information-based features. The optimal feature set was identified from each encoding, and their respective models were established. Extensive experimental results show that iRice-MS always display excellent performance on 5-fold cross-validation and independent dataset test. In addition, our novel approach provides the superiority to other existing tools in terms of AUC value. Based on the proposed model, a web server named iRice-MS was established and is freely accessible at http://lin-group.cn/server/iRice-MS.

Ferroptosis Involved in Cardiovascular Diseases: Mechanism Exploration of Ferroptosis' Role in Common Pathological Changes.

ABSTRACT: Regulated cell death is a controlled form of cell death that protects cells by adaptive responses in pathophysiological states. Ferroptosis has been identified as a novel method of controlling cell death in recent years. Several cardiovascular diseases (CVDs) are shown to be profoundly influenced by ferroptosis, and ferroptosis is directly linked to the majority of cardiovascular pathological alterations. Despite this, it is still unclear how ferroptosis affects the pathogenic alterations that take place in CVDs. Based on a review of the mechanisms that regulate ferroptosis, this review explores the most recent research on the role of ferroptosis in the major pathological changes associated with CVDs, to provide new perspectives and strategies for cardiovascular research and clinical treatment.

E3 ubiquitin ligase NEDD4L inhibits epithelial-mesenchymal transition by suppressing the ß-catenin/HIF-1α positive feedback loop in chronic rhinosinusitis with nasal polyps.

Chronic rhinosinusitis with nasal polyp (CRSwNP) is a refractory inflammatory disease with epithelial-mesenchymal transition (EMT) as one of the key features. Since ubiquitin modification has been shown to regulate the EMT process in other diseases, targeting ubiquitin ligases may be a potential strategy for the treatment of CRSwNP. In this study we investigated whether certain E3 ubiquitin ligases could regulate the EMT process in CRSwNP, and whether these regulations could be the potential drug targets as well as the underlying mechanisms. After screening the potential drug target by bioinformatic analyses, the expression levels of three potential E3 ubiquitin ligases were compared among the control, eosinophilic nasal polyp (ENP) and non-eosinophilic nasal polyp (NENP) group in clinical samples, and the significant decrement of the expression level of NEDD4L was found. Then, IP-MS, bioinformatics and immunohistochemistry studies suggested that low NEDD4L expression may be associated with the EMT process. In human nasal epithelial cells (hNECs) and human nasal epithelial cell line RPMI 2650, knockdown of NEDD4L promoted EMT, while upregulating NEDD4L reversed this effect, suggesting that NEDD4L inhibited EMT in nasal epithelial cells. IP-MS and Co-IP studies revealed that NEDD4L mediated the degradation of DDR1. We demonstrated that NEDD4L inhibited the ß-catenin/HIF-1α positive feedback loop either directly (degrading ß-catenin and HIF-1α) or indirectly (mediating DDR1 degradation). These results were confirmed in a murine NP model in vivo. This study for the first time reveals the regulatory role of ubiquitin in the EMT process of nasal epithelial cells, and identifies a novel drug target NEDD4L, which has promising efficacy against both ENP and NENP by suppressing ß-catenin/HIF-1α positive feedback loop.

Using part of the initial analgesic dose as the epidural test dose did not delay the onset of labor analgesia: a randomized controlled clinical trial.

BACKGROUND: Epidural test dose for labor analgesia is controversial and varies widely in clinical practice. It is currently unclear whether using a portion of the initial dose for analgesia as the test dose delays the onset time of analgesia, compared to the traditional test dose. METHODS: One hundred and twenty-six parturients who chose epidural analgesia during labor were randomly assigned to two groups. The first dose in group L was 3 ml 1.5% lidocaine, and in the RF group was 10 ml 0.1% ropivacaine combined with 2 µg/ml fentanyl. After 3 min of observation, both groups received 8 ml 0.1% ropivacaine combined with 2 µg/ml fentanyl. The onset time of analgesia, motor and sensory blockade level, numerical pain rating scale, patient satisfaction score, and side effects were recorded. RESULTS: The onset time of analgesia in group RF was similar to that in group L (group RF vs group L, 7.0 [5.0-9.0] minutes vs 8.0 [5.0-11.0] minutes, p = 0.197). The incidence of foot numbness (group RF vs group L, 34.9% vs 57.1%, p = 0.020) and foot warming (group RF vs group L, 15.9% vs 47.6%, p < 0.001) in group RF was significantly lower than that in group L. There was no difference between the two groups on other outcomes. CONCLUSIONS: Compared with 1.5% lidocaine 3 ml, 0.1% ropivacaine 10 ml combined with 2 µg/ml fentanyl as an epidural test dose did not delay the onset of labor analgesia, and the side effects were slightly reduced. CLINICAL TRIAL REGISTRATION: http://www.chictr.org.cn (ChiCTR2100043071).

Probing the Effect of Alloying Elements on the Interfacial Segregation Behavior and Electronic Properties of Mg/Ti Interface via First-Principles Calculations.

The interface connects the reinforced phase and the matrix of materials, with its microstructure and interfacial configurations directly impacting the overall performance of composites. In this study, utilizing seven atomic layers of Mg(0001) and Ti(0001) surface slab models, four different Mg(0001)/Ti(0001) interfaces with varying atomic stacking configurations were constructed. The calculated interface adhesion energy and electronic bonding information of the Mg(0001)/Ti(0001) interface reveal that the HCP2 interface configuration exhibits the best stability. Moreover, Si, Ca, Sc, V, Cr, Mn, Fe, Cu, Zn, Y, Zr, Nb, Mo, Sn, La, Ce, Nd, and Gd elements are introduced into the Mg/Ti interface layer or interfacial sublayer of the HCP2 configurations, and their interfacial segregation behavior is investigated systematically. The results indicate that Gd atom doping in the Mg(0001)/Ti(0001) interface exhibits the smallest heat of segregation, with a value of -5.83 eV. However, Ca and La atom doping in the Mg(0001)/Ti(0001) interface show larger heat of segregation, with values of 0.84 and 0.63 eV, respectively. This implies that the Gd atom exhibits a higher propensity to segregate at the interface, whereas the Ca and La atoms are less inclined to segregate. Moreover, the electronic density is thoroughly analyzed to elucidate the interfacial segregation behavior. The research findings presented in this paper offer valuable guidance and insights for designing the composition of magnesium-based composites.

Dissecting Causal Relationships Between Plasma Metabolites and Osteoporosis: A Bidirectional Mendelian Randomization Study.

Objective To investigate the causal relationships between plasma metabolites and osteoporosis via Mendelian randomization (MR) analysis. Methods Bidirectional MR was used to analyze pooled data from different genome-wide association studies (GWAS) to investigate the causal relationships between plasma metabolites and osteoporosis. The causal effect of plasma metabolites on osteoporosis was estimated using the inverse variance weighted method, intersections of statistically significant metabolites obtained from different sources of osteoporosis-related GWAS aggregated data was determined, and then sensitivity analysis was performed on these metabolites. Heterogeneity between single nucleotide polymorphisms was evaluated by Cochran's Q test. Horizontal pleiotropy was assessed through the application of the MR-Egger intercept method and the MR-PRESSO method. The causal effect of osteoporosis on plasma metabolites was also evaluated using the inverse variance weighted method. Additionally, pathway analysis was conducted to identify potential metabolic pathways involved in the regulation of osteoporosis. Results After primary analysis and a series of sensitivity analyses, 77 and 61 plasma metabolites were identified as having a causal relationship with osteoporosis from the GWAS data in the GCST90038656 and GCST90044600 datasets , respectively. Five common metabolites were identified via intersection. X-13684 levels (GCST90038656: OR = 0.999, 95% CI, 0.998-1.000, P = 0.004; GCST90044600 (OR = 0.834, 95% CI, 0.700-0.993, P = 0.042), and the glucose-to-maltose ratio (GCST90038656: OR = 0.998, 95% CI, 0.997-1.000, P = 0.025; GCST90044600: OR = 0.752, 95% CI, 0.576-0.981, P = 0.036) were negatively associated with osteoporosis, whereas glycoursodeoxycholate levels (GCST90038656: OR = 1.002, 95% CI, 1.000-1.003, P = 0.032; GCST90044600: OR = 1.331, 95% CI, 1.036-1.709, P = 0.025) and arachidoylcarnitine (C20) levels (GCST90038656: OR = 1.001, 95% CI, 1.000-1.003, P = 0.039; GCST90044600: OR = 1.237; 95% CI, 1.008-1.518, P = 0.042) were positively associated with osteoporosis. The relationship between X-11299 levels and osteoporosis showed contradictory results (GCST90038656: OR= 0.998, 95% CI, 0.997-1.000, P = 0.026; GCST90044600: OR = 1.402, 95% CI, 1.071-1.834, P = 0.014). Pathway analysis indicated that glycine, serine, and threonine metabolism, valine, leucine, and isoleucine biosynthesis, galactose metabolism, arginine biosynthesis, and starch and sucrose metabolism pathways were participated in the development of osteoporosis. Conclusion We found a causal relationship between plasma metabolites and osteoporosis. These results offer novel perspectives that have implications for targeted interventions focused on metabolites in the management of osteoporosis.

LPS-TLR4 pathway exaggerates alcoholic hepatitis via provoking NETs formation.

BACKGROUND: Intrahepatic infiltration of neutrophils is a character of alcoholic hepatitis (AH) and neutrophil extracellular traps (NETs) are an important strategy for neutrophils to fix and kill invading microorganisms. The gut-liver axis has been thought to play a critical role in many liver diseases also including AH. However, whether NETs appear in AH and play role in AH is still unsure. METHODS: Serum samples from AH patients were collected and LPS and MPO-DNA were detected. WT, NE KO, and TLR4 KO mice were used to build the AH model, and the intestinal bacteria were eliminated at the same time and LPS was given. Then the formation of NETs and AH-related markers were detected. RESULTS: The serum MPO-DNA and LPS concentration was increased in AH patients and a correlation was revealed between these two indexes. More intrahepatic NETs formed in AH mice. NETs formation decreased with antibiotic intervention and restored with antibiotic intervention plus LPS supplement. While NETs formation failed to change with gut microbiome or combine LPS supplement in TLR4 KO mice. As we tested AH-related characters, liver injury, intrahepatic fat deposition, inflammation, and fibrosis alleviated with depletion of NE. These related marks were also attenuated with gut sterilization by antibiotics and recovered with a combined treatment with antibiotics plus LPS. But the AH-related markers did show a difference in TLR4 KO mice when they received the same treatment. CONCLUSION: Intestinal-derived LPS promotes NETs formation in AH through the TLR4 pathway and further accelerates the AH process by NETs.

Ir-Doped CuPd Single-Crystalline Mesoporous Nanotetrahedrons for Ethylene Glycol Oxidation Electrocatalysis: Enhanced Selective Cleavage of C-C Bond.

The development of highly efficient electrocatalysts for complete oxidation of ethylene glycol (EG) in direct EG fuel cells is of decisive importance to hold higher energy efficiency. Despite some achievements, their progress, especially electrocatalytic selectivity to complete oxidated C1 products, is remarkably slower than expected. In this work, we developed a facile aqueous synthesis of Ir-doped CuPd single-crystalline mesoporous nanotetrahedrons (Ir-CuPd SMTs) as high-performance electrocatalyst for promoting oxidation cleavage of C-C bond in alkaline EG oxidation reaction (EGOR) electrocatalysis. The synthesis relied on precise reduction/co-nucleation and epitaxial growth of Ir, Cu and Pd precursors with cetyltrimethylammonium chloride as the mesopore-forming surfactant and extra Br- as the facet-selective agent under ambient conditions. The products featured concave nanotetrahedron morphology enclosed by well-defined (111) facets, single-crystalline and mesoporous structure radiated from the center, and uniform elemental composition without any phase separation. Ir-CuPd SMTs disclosed remarkably enhanced electrocatalytic activity and excellent stability as well as superior selectivity of C1 products for alkaline EGOR electrocatalysis. Detailed mechanism studies demonstrated that performance improvement came from structural and compositional synergies, which kinetically accelerated transports of electrons/reactants within active sites of penetrated mesopores and facilitated oxidation cleavage of high-energy-barrier C-C bond of EG for desired C1 products. More interestingly, Ir-CuPd SMTs performed well in coupled electrocatalysis of anode EGOR and cathode nitrate reduction, highlighting its high potential as bifunctional electrocatalyst in various applications.

TET2 deficiency promotes anxiety and depression-like behaviors by activating NLRP3/IL-1ß pathway in microglia of allergic rhinitis mice.

BACKGROUND: Anxiety and depression-like behaviors in allergic rhinitis (AR) are attracting attention, while the precise mechanism has not been clearly elucidated. Recent evidence shows that neuroinflammation in anterior cingulate cortex (ACC) may be the core of these neuropsychiatric symptoms in AR. Here, we investigated the molecular link between the anxiety and depression-like behaviors and neuroinflammation in ACC. METHODS: Mice were sensitized and challenged with ovalbumin (OVA) to induce AR. Nasal inflammation levels were assessed by H&E staining and PAS staining. Anxiety and depression-like behaviors were evaluated by behavioral experiments including open field test, forced swimming test, and sucrose preference test. Neuronal impairment was characterized via Nissl staining and 18FDG-PET. The role of ten-eleven translocation 2 (TET2) in AR-related anxiety and depression was assessed by Tet2-/- mice. In addition, the murine BV2 microglial cell line was utilized to explore the molecular mechanisms by which TET2 mediates neuroinflammation. The levels of TET2, NLRP3 and their downstream molecules were detected by immunohistochemistry, Western blot, Dot blot and ELISA. The effects of metformin on depression-like behaviors in AR mice were also evaluated. RESULTS: AR mice showed significant anxiety and depression-like behaviors, which associated with the activation of ACC. Loss of TET2 activated the NLRP3/IL-1ß pathway of microglia in AR mice, further accelerating the anxiety and depression-like behaviors. In addition, knockdown of TET2 activated the NLRP3/IL-1ß pathway in BV2 cells. Metformin improved the neuropsychiatric symptoms of AR mice by reducing the activation of NLRP3/IL-1ß pathway after upregulating TET2. CONCLUSION: TET2 deficiency activates the NLRP3/IL-1ß pathway of microglia in the ACC, promoting the pathological process of anxiety and depression-like behavior in AR. Metformin could be effective in treating neuroinflammation by regulating microglia via TET2 up-regulation, indicating that metformin is a potential way to treat anxiety and depression-like behaviors in AR.

DeepIPs: comprehensive assessment and computational identification of phosphorylation sites of SARS-CoV-2 infection using a deep learning-based approach.

The rapid spread of SARS-CoV-2 infection around the globe has caused a massive health and socioeconomic crisis. Identification of phosphorylation sites is an important step for understanding the molecular mechanisms of SARS-CoV-2 infection and the changes within the host cells pathways. In this study, we present DeepIPs, a first specific deep-learning architecture to identify phosphorylation sites in host cells infected with SARS-CoV-2. DeepIPs consists of the most popular word embedding method and convolutional neural network-long short-term memory network architecture to make the final prediction. The independent test demonstrates that DeepIPs improves the prediction performance compared with other existing tools for general phosphorylation sites prediction. Based on the proposed model, a web-server called DeepIPs was established and is freely accessible at http://lin-group.cn/server/DeepIPs. The source code of DeepIPs is freely available at the repository https://github.com/linDing-group/DeepIPs.

Deep-Kcr: accurate detection of lysine crotonylation sites using deep learning method.

As a newly discovered protein posttranslational modification, histone lysine crotonylation (Kcr) involved in cellular regulation and human diseases. Various proteomics technologies have been developed to detect Kcr sites. However, experimental approaches for identifying Kcr sites are often time-consuming and labor-intensive, which is difficult to widely popularize in large-scale species. Computational approaches are cost-effective and can be used in a high-throughput manner to generate relatively precise identification. In this study, we develop a deep learning-based method termed as Deep-Kcr for Kcr sites prediction by combining sequence-based features, physicochemical property-based features and numerical space-derived information with information gain feature selection. We investigate the performances of convolutional neural network (CNN) and five commonly used classifiers (long short-term memory network, random forest, LogitBoost, naive Bayes and logistic regression) using 10-fold cross-validation and independent set test. Results show that CNN could always display the best performance with high computational efficiency on large dataset. We also compare the Deep-Kcr with other existing tools to demonstrate the excellent predictive power and robustness of our method. Based on the proposed model, a webserver called Deep-Kcr was established and is freely accessible at http://lin-group.cn/server/Deep-Kcr.

A computational platform to identify origins of replication sites in eukaryotes.

The locations of the initiation of genomic DNA replication are defined as origins of replication sites (ORIs), which regulate the onset of DNA replication and play significant roles in the DNA replication process. The study of ORIs is essential for understanding the cell-division cycle and gene expression regulation. Accurate identification of ORIs will provide important clues for DNA replication research and drug development by developing computational methods. In this paper, the first integrated predictor named iORI-Euk was built to identify ORIs in multiple eukaryotes and multiple cell types. In the predictor, seven eukaryotic (Homo sapiens, Mus musculus, Drosophila melanogaster, Arabidopsis thaliana, Pichia pastoris, Schizosaccharomyces pombe and Kluyveromyces lactis) ORI data was collected from public database to construct benchmark datasets. Subsequently, three feature extraction strategies which are k-mer, binary encoding and combination of k-mer and binary were used to formulate DNA sequence samples. We also compared the different classification algorithms' performance. As a result, the best results were obtained by using support vector machine in 5-fold cross-validation test and independent dataset test. Based on the optimal model, an online web server called iORI-Euk (http://lin-group.cn/server/iORI-Euk/) was established for the novel ORI identification.

iDHS-Deep: an integrated tool for predicting DNase I hypersensitive sites by deep neural network.

DNase I hypersensitive site (DHS) refers to the hypersensitive region of chromatin for the DNase I enzyme. It is an important part of the noncoding region and contains a variety of regulatory elements, such as promoter, enhancer, and transcription factor-binding site, etc. Moreover, the related locus of disease (or trait) are usually enriched in the DHS regions. Therefore, the detection of DHS region is of great significance. In this study, we develop a deep learning-based algorithm to identify whether an unknown sequence region would be potential DHS. The proposed method showed high prediction performance on both training datasets and independent datasets in different cell types and developmental stages, demonstrating that the method has excellent superiority in the identification of DHSs. Furthermore, for the convenience of related wet-experimental researchers, the user-friendly web-server iDHS-Deep was established at http://lin-group.cn/server/iDHS-Deep/, by which users can easily distinguish DHS and non-DHS and obtain the corresponding developmental stage ofDHS.

A sequence-based deep learning approach to predict CTCF-mediated chromatin loop.

Three-dimensional (3D) architecture of the chromosomes is of crucial importance for transcription regulation and DNA replication. Various high-throughput chromosome conformation capture-based methods have revealed that CTCF-mediated chromatin loops are a major component of 3D architecture. However, CTCF-mediated chromatin loops are cell type specific, and most chromatin interaction capture techniques are time-consuming and labor-intensive, which restricts their usage on a very large number of cell types. Genomic sequence-based computational models are sophisticated enough to capture important features of chromatin architecture and help to identify chromatin loops. In this work, we develop Deep-loop, a convolutional neural network model, to integrate k-tuple nucleotide frequency component, nucleotide pair spectrum encoding, position conservation, position scoring function and natural vector features for the prediction of chromatin loops. By a series of examination based on cross-validation, Deep-loop shows excellent performance in the identification of the chromatin loops from different cell types. The source code of Deep-loop is freely available at the repository https://github.com/linDing-group/Deep-loop.

DeepYY1: a deep learning approach to identify YY1-mediated chromatin loops.

The protein Yin Yang 1 (YY1) could form dimers that facilitate the interaction between active enhancers and promoter-proximal elements. YY1-mediated enhancer-promoter interaction is the general feature of mammalian gene control. Recently, some computational methods have been developed to characterize the interactions between DNA elements by elucidating important features of chromatin folding; however, no computational methods have been developed for identifying the YY1-mediated chromatin loops. In this study, we developed a deep learning algorithm named DeepYY1 based on word2vec to determine whether a pair of YY1 motifs would form a loop. The proposed models showed a high prediction performance (AUCs$\ge$0.93) on both training datasets and testing datasets in different cell types, demonstrating that DeepYY1 has an excellent performance in the identification of the YY1-mediated chromatin loops. Our study also suggested that sequences play an important role in the formation of YY1-mediated chromatin loops. Furthermore, we briefly discussed the distribution of the replication origin site in the loops. Finally, a user-friendly web server was established, and it can be freely accessed at http://lin-group.cn/server/DeepYY1.

Application of artificial intelligence and machine learning for COVID-19 drug discovery and vaccine design.

The global pandemic of coronavirus disease 2019 (COVID-19), caused by severe acute respiratory syndrome coronavirus 2, has led to a dramatic loss of human life worldwide. Despite many efforts, the development of effective drugs and vaccines for this novel virus will take considerable time. Artificial intelligence (AI) and machine learning (ML) offer promising solutions that could accelerate the discovery and optimization of new antivirals. Motivated by this, in this paper, we present an extensive survey on the application of AI and ML for combating COVID-19 based on the rapidly emerging literature. Particularly, we point out the challenges and future directions associated with state-of-the-art solutions to effectively control the COVID-19 pandemic. We hope that this review provides researchers with new insights into the ways AI and ML fight and have fought the COVID-19 outbreak.