MLSNet: a deep learning model for predicting transcription factor binding sites.

Accurate prediction of transcription factor binding sites (TFBSs) is essential for understanding gene regulation mechanisms and the etiology of diseases. Despite numerous advances in deep learning for predicting TFBSs, their performance can still be enhanced. In this study, we propose MLSNet, a novel deep learning architecture designed specifically to predict TFBSs. MLSNet innovatively integrates multisize convolutional fusion with long short-term memory (LSTM) networks to effectively capture DNA-sparse higher-order sequence features. Further, MLSNet incorporates super token attention and Bi-LSTM to systematically extract and integrate higher-order DNA shape features. Experimental results on 165 ChIP-seq (chromatin immunoprecipitation followed by sequencing) datasets indicate that MLSNet consistently outperforms several state-of-the-art algorithms in the prediction of TFBSs. Specifically, MLSNet reports average metrics: 0.8306 for ACC, 0.8992 for AUROC, and 0.9035 for AUPRC, surpassing the second-best methods by 1.82%, 1.68%, and 1.54%, respectively. This research delineates the effectiveness of combining multi-size convolutional layers with LSTM and DNA shape-based features in enhancing predictive accuracy. Moreover, this study comprehensively assesses the variability in model performance across different cell lines and transcription factors. The source code of MLSNet is available at https://github.com/minghaidea/MLSNet.

Does Microbiome Contribute to Longevity? Compositional and Functional Differences in Gut Microbiota in Chinese Long-Living (>90 Years) and Elderly (65-74 Years) Adults.

The study of longevity and its determinants has been revitalized with the rise of microbiome scholarship. The gut microbiota have been established to play essential protective, metabolic, and physiological roles in human health and disease. The gut dysbiosis has been identified as an important factor contributing to the development of multiple diseases. Accordingly, it is reasonable to hypothesize that the gut microbiota of long-living individuals have healthy antiaging-associated gut microbes, which, by extension, might provide specific molecular targets for antiaging treatments and interventions. In the present study, we compared the gut microbiota of Chinese individuals in two different age groups, long-living adults (aged over 90 years) and elderly adults (aged 65-74 years) who were free of major diseases. We found significantly lower relative abundances of bacteria in the genera Sutterella and Megamonas in the long-living individuals. Furthermore, we established that while biological processes such as autophagy (GO:0006914) and telomere maintenance through semiconservative replication (GO:0032201) were enhanced in the long-living group, response to lipopolysaccharide (GO:0032496), nicotinamide adenine dinucleotide oxidation (GO:0006116), and S-adenosyl methionine metabolism (GO:0046500) were weakened. Moreover, the two groups were found to differ with respect to amino acid metabolism. We suggest that these compositional and functional differences in the gut microbiota may potentially be associated with mechanisms that contribute to determining longevity or aging.

MetalTrans: A Biological Language Model-Based Approach for Predicting Disease-Associated Mutations in Protein Metal-Binding Sites.

The critical importance of accurately predicting mutations in protein metal-binding sites for advancing drug discovery and enhancing disease diagnostic processes cannot be overstated. In response to this imperative, MetalTrans emerges as an accurate predictor for disease-associated mutations in protein metal-binding sites. The core innovation of MetalTrans lies in its seamless integration of multifeature splicing with the Transformer framework, a strategy that ensures exhaustive feature extraction. Central to MetalTrans's effectiveness is its deep feature combination strategy, which merges evolutionary-scale modeling amino acid embeddings with ProtTrans embeddings, thus shedding light on the biochemical properties of proteins. Employing the Transformer component, MetalTrans leverages the self-attention mechanism to delve into higher-level representations. Utilizing mutation site information for feature fusion not only enriches the feature set but also sidesteps the common pitfall of overestimation linked to protein sequence-based predictions. This nuanced approach to feature fusion is a key differentiator, enabling MetalTrans to outperform existing methods significantly, as evidenced by comparative analyses. Our evaluations across varied metal binding site data sets (specifically Zn, Ca, Mg, and Mix) underscore MetalTrans's superior performance, which achieved the average AUC values of 0.971, 0.965, 0.980, and 0.945 on multiple 5-fold cross-validation, respectively. Remarkably, against the multichannel convolutional neural network method on a benchmark independent test set, MetalTrans demonstrated unparalleled robustness and superiority, boasting the AUC score of 0.998 on multiple 5-fold cross-validation. Our comprehensive examination of the predicted outcomes further confirms the effectiveness of the model. The source codes, data sets, and prediction results for MetalTrans can be accessed for academic usage at https://github.com/EduardWang/MetalTrans.

Application of m6A regulators to predict transformation from myelodysplastic syndrome to acute myeloid leukemia via machine learning.

Myelodysplastic syndrome (MDS) frequently transforms into acute myeloid leukemia (AML). Predicting the risk of its transformation will help to make the treatment plan. Levels of expression of N6-methyladenosine (m6A) regulators is difference in patients with AML, MDS, and MDS transformed into AML. Seven machine learning algorithms were established based on all of 26 m6A or main differentially expressed m6A regulator genes, and attempted to establish a risk assessment method to distinguish AML from MDS and predict the transformation of MDS into AML. In collective of m6A regulators sets, support vector machine (SVM) and neural network (NNK) model best distinguished AML or MDS from control, with area under the ROC curve (AUROC) 0.966 and 0.785 respectively. The SVM model best distinguished MDS from AML, with AUROC 0.943, sensitivity 0.862, specificity 0.864, and accuracy 0.864. In differentially expressed gene sets, SVM and logistic regression (LR) model best distinguished AML or MDS from control, with AUROC 0.945 and 0.801 respectively. The random forest (RF) model best distinguished between MDS and AML, with AUROC 0.928, sensitivity 0.725, specificity 0.898, and accuracy 0.818. For predictive capacity of MDS transformed into AML, SVM model showed the best predicted in collective m6A regulators sets, with AUROC 0.781 and accuracy 0.740. The LR model showed the best predicted in differential expression m6A regulators sets, with AUROC 0.820 and accuracy 0.760. All results suggested that machine learning model established by m6A regulators can be used to distinguished AML or MDS from control, distinguished AML from MDS and predicted the transformation of MDS into AML.

Evaluation of enzyme activity predictions for variants of unknown significance in Arylsulfatase A.

Continued advances in variant effect prediction are necessary to demonstrate the ability of machine learning methods to accurately determine the clinical impact of variants of unknown significance (VUS). Towards this goal, the ARSA Critical Assessment of Genome Interpretation (CAGI) challenge was designed to characterize progress by utilizing 219 experimentally assayed missense VUS in the Arylsulfatase A (ARSA) gene to assess the performance of community-submitted predictions of variant functional effects. The challenge involved 15 teams, and evaluated additional predictions from established and recently released models. Notably, a model developed by participants of a genetics and coding bootcamp, trained with standard machine-learning tools in Python, demonstrated superior performance among submissions. Furthermore, the study observed that state-of-the-art deep learning methods provided small but statistically significant improvement in predictive performance compared to less elaborate techniques. These findings underscore the utility of variant effect prediction, and the potential for models trained with modest resources to accurately classify VUS in genetic and clinical research.

Enantioselective Nickel-Catalyzed Denitrogenative Transannulation En Route to N-N Atropisomers.

Nickel-catalyzed transannulation reactions triggered by the extrusion of small gaseous molecules have emerged as a powerful strategy for the efficient construction of heterocyclic compounds. However, their use in asymmetric synthesis remains challenging because of the difficulty in controlling stereo- and regioselectivity. Herein, we report the first nickel-catalyzed asymmetric synthesis of N-N atropisomers by the denitrogenative transannulation of benzotriazones with alkynes. A broad range of N-N atropisomers was obtained with excellent regio- and enantioselectivity under mild conditions. Moreover, density functional theory (DFT) calculations provided insights into the nickel-catalyzed reaction mechanism and enantioselectivity control.

Prediction of protein-ATP binding residues using multi-view feature learning via contextual-based co-attention network.

Accurately predicting protein-ATP binding residues is critical for protein function annotation and drug discovery. Computational methods dedicated to the prediction of binding residues based on protein sequence information have exhibited notable advancements in predictive accuracy. Nevertheless, these methods continue to grapple with several formidable challenges, including limited means of extracting more discriminative features and inadequate algorithms for integrating protein and residue information. To address the problems, we propose ATP-Deep, a novel protein-ATP binding residues predictor. ATP-Deep harnesses the capabilities of unsupervised pre-trained language models and incorporates domain-specific evolutionary context information from homologous sequences. It further refines the embedding at the residue level through integration with corresponding protein-level information and employs a contextual-based co-attention mechanism to adeptly fuse multiple sources of features. The performance evaluation results on the benchmark datasets reveal that ATP-Deep achieves an AUC of 0.954 and 0.951, respectively, surpassing the performance of the state-of-the-art model. These findings underscore the effectiveness of assimilating protein-level information and deploying a contextual-based co-attention mechanism grounded in context to bolster the prediction performance of protein-ATP binding residues.

TransEFVP: A Two-Stage Approach for the Prediction of Human Pathogenic Variants Based on Protein Sequence Embedding Fusion.

Studying the effect of single amino acid variations (SAVs) on protein structure and function is integral to advancing our understanding of molecular processes, evolutionary biology, and disease mechanisms. Screening for deleterious variants is one of the crucial issues in precision medicine. Here, we propose a novel computational approach, TransEFVP, based on large-scale protein language model embeddings and a transformer-based neural network to predict disease-associated SAVs. The model adopts a two-stage architecture: the first stage is designed to fuse different feature embeddings through a transformer encoder. In the second stage, a support vector machine model is employed to quantify the pathogenicity of SAVs after dimensionality reduction. The prediction performance of TransEFVP on blind test data achieves a Matthews correlation coefficient of 0.751, an F1-score of 0.846, and an area under the receiver operating characteristic curve of 0.871, higher than the existing state-of-the-art methods. The benchmark results demonstrate that TransEFVP can be explored as an accurate and effective SAV pathogenicity prediction method. The data and codes for TransEFVP are available at https://github.com/yzh9607/TransEFVP/tree/master for academic use.

FCMSTrans: Accurate Prediction of Disease-Associated nsSNPs by Utilizing Multiscale Convolution and Deep Feature Combination within a Transformer Framework.

Nonsynonymous single-nucleotide polymorphisms (nsSNPs), implicated in over 6000 diseases, necessitate accurate prediction for expedited drug discovery and improved disease diagnosis. In this study, we propose FCMSTrans, a novel nsSNP predictor that innovatively combines the transformer framework and multiscale modules for comprehensive feature extraction. The distinctive attribute of FCMSTrans resides in a deep feature combination strategy. This strategy amalgamates evolutionary-scale modeling (ESM) and ProtTrans (PT) features, providing an understanding of protein biochemical properties, and position-specific scoring matrix, secondary structure, predicted relative solvent accessibility, and predicted disorder (PSPP) features, which are derived from four protein sequences and structure-oriented characteristics. This feature combination offers a comprehensive view of the molecular dynamics involving nsSNPs. Our model employs the transformer's self-attention mechanisms across multiple layers, extracting higher-level and abstract representations. Simultaneously, varied-level features are captured by multiscale convolutions, enriching feature abstraction at multiple echelons. Our comparative analyses with existing methodologies highlight significant improvements made possible by the integrated feature fusion approach adopted in FCMSTrans. This is further substantiated by performance assessments based on diverse data sets, such as PredictSNP, MMP, and PMD, with areas under the curve (AUCs) of 0.869, 0.819, and 0.693, respectively. Furthermore, FCMSTrans shows robustness and superiority by outperforming the current best predictor, PROVEAN, in a blind test conducted on a third-party data set, achieving an impressive AUC score of 0.7838. The Python code of FCMSTrans is available at https://github.com/gc212/FCMSTrans for academic usage.

Review of Computational Methods and Database Sources for Predicting the Effects of Coding Frameshift Small Insertion and Deletion Variations.

Genetic variations (including substitutions, insertions, and deletions) exert a profound influence on DNA sequences. These variations are systematically classified as synonymous, nonsynonymous, and nonsense, each manifesting distinct effects on proteins. The implementation of high-throughput sequencing has significantly augmented our comprehension of the intricate interplay between gene variations and protein structure and function, as well as their ramifications in the context of diseases. Frameshift variations, particularly small insertions and deletions (indels), disrupt protein coding and are instrumental in disease pathogenesis. This review presents a succinct review of computational methods, databases, current challenges, and future directions in predicting the consequences of coding frameshift small indels variations. We analyzed the predictive efficacy, reliability, and utilization of computational methods and variant account, reliability, and utilization of database. Besides, we also compared the prediction methodologies on GOF/LOF pathogenic variation data. Addressing the challenges pertaining to prediction accuracy and cross-species generalizability, nascent technologies such as AI and deep learning harbor immense potential to enhance predictive capabilities. The importance of interdisciplinary research and collaboration cannot be overstated for devising effective diagnosis, treatment, and prevention strategies concerning diseases associated with coding frameshift indels variations.

Superelastic and superflexible cellulose aerogels for thermal insulation and oil/water separation.

Aerogels with low thermal conductivity and high adsorption capacity present a promising solution to curb water pollution caused by organic reagents as well as mitigate heat loss. Although aerogels exhibiting good adsorption capacity and thermal insulation have been reported, materials with mechanical integrity, high flexibility and shear resistance still pose a formidable task. Here, we produced bacterial cellulose-based ultralight multifunctional hybrid aerogels by using freeze-drying followed by chemical vapor deposition silylation method. The hybrid aerogels displayed a low density of 10-15 mg/cm3, high porosity exceeding 99.1 %, low thermal conductivity (27.3-29.2 mW/m.K) and superior hydrophobicity (water contact angle>120o). They also exhibited excellent mechanical properties including superelasticity, high flexibility and shear resistance. The hybrid aerogels demonstrated high heat shielding efficiency when used as an insulating material. As a selective oil absorbent, the hybrid aerogels exhibit a maximum adsorption capacity of up to approximately 156 times its own weight and excellent recoverability. Especially, the aerogel's highly accessible porous microstructure results in an impressive flux rate of up to 162 L/h.g when used as a filter in a continuous oil-water separator to isolate n-hexane-water mixtures. This work presents a novel endeavor to create high-performance, sustainable, reusable, and adaptable multifunctional aerogels.

Risk of thrombosis and bleeding in gynecologic cancer surgery: systematic review and meta-analysis.

OBJECTIVE: This study aimed to provide procedure-specific estimates of the risk of symptomatic venous thromboembolism and major bleeding in the absence of thromboprophylaxis, following gynecologic cancer surgery. DATA SOURCES: We conducted comprehensive searches on Embase, MEDLINE, Web of Science, and Google Scholar for observational studies. We also reviewed reference lists of eligible studies and review articles. We performed separate searches for randomized trials addressing effects of thromboprophylaxis and conducted a web-based survey on thromboprophylaxis practice. STUDY ELIGIBILITY CRITERIA: Observational studies enrolling ≥50 adult patients undergoing gynecologic cancer surgery procedures reporting absolute incidence for at least 1 of the following were included: symptomatic pulmonary embolism, symptomatic deep vein thrombosis, symptomatic venous thromboembolism, bleeding requiring reintervention (including reexploration and angioembolization), bleeding leading to transfusion, or postoperative hemoglobin <70 g/L. METHODS: Two reviewers independently assessed eligibility, performed data extraction, and evaluated risk of bias of eligible articles. We adjusted the reported estimates for thromboprophylaxis and length of follow-up and used the median value from studies to determine cumulative incidence at 4 weeks postsurgery stratified by patient venous thromboembolism risk factors. The GRADE approach was applied to rate evidence certainty. RESULTS: We included 188 studies (398,167 patients) reporting on 37 gynecologic cancer surgery procedures. The evidence certainty was generally low to very low. Median symptomatic venous thromboembolism risk (in the absence of prophylaxis) was <1% in 13 of 37 (35%) procedures, 1% to 2% in 11 of 37 (30%), and >2.0% in 13 of 37 (35%). The risks of venous thromboembolism varied from 0.1% in low venous thromboembolism risk patients undergoing cervical conization to 33.5% in high venous thromboembolism risk patients undergoing pelvic exenteration. Estimates of bleeding requiring reintervention varied from <0.1% to 1.3%. Median risks of bleeding requiring reintervention were <1% in 22 of 29 (76%) and 1% to 2% in 7 of 29 (24%) procedures. CONCLUSION: Venous thromboembolism reduction with thromboprophylaxis likely outweighs the increase in bleeding requiring reintervention in many gynecologic cancer procedures (eg, open surgery for ovarian cancer and pelvic exenteration). In some procedures (eg, laparoscopic total hysterectomy without lymphadenectomy), thromboembolism and bleeding risks are similar, and decisions depend on individual risk prediction and values and preferences regarding venous thromboembolism and bleeding.

Risk of thrombosis and bleeding in gynecologic noncancer surgery: systematic review and meta-analysis.

OBJECTIVE: This study aimed to provide procedure-specific estimates of the risk for symptomatic venous thromboembolism and major bleeding in noncancer gynecologic surgeries. DATA SOURCES: We conducted comprehensive searches on Embase, MEDLINE, Web of Science, and Google Scholar. Furthermore, we performed separate searches for randomized trials that addressed the effects of thromboprophylaxis. STUDY ELIGIBILITY CRITERIA: Eligible studies were observational studies that enrolled ≥50 adult patients who underwent noncancer gynecologic surgery procedures and that reported the absolute incidence of at least 1 of the following: symptomatic pulmonary embolism, symptomatic deep vein thrombosis, symptomatic venous thromboembolism, bleeding that required reintervention (including re-exploration and angioembolization), bleeding that led to transfusion, or postoperative hemoglobin level <70 g/L. METHODS: A teams of 2 reviewers independently assessed eligibility, performed data extraction, and evaluated the risk of bias of the eligible articles. We adjusted the reported estimates for thromboprophylaxis and length of follow-up and used the median value from studies to determine the cumulative incidence at 4 weeks postsurgery stratified by patient venous thromboembolism risk factors and used the Grading of Recommendations Assessment, Development and Evaluation approach to rate the evidence certainty. RESULTS: We included 131 studies (1,741,519 patients) that reported venous thromboembolism risk estimates for 50 gynecologic noncancer procedures and bleeding requiring reintervention estimates for 35 procedures. The evidence certainty was generally moderate or low for venous thromboembolism and low or very low for bleeding requiring reintervention. The risk for symptomatic venous thromboembolism varied from a median of <0.1% for several procedures (eg, transvaginal oocyte retrieval) to 1.5% for others (eg, minimally invasive sacrocolpopexy with hysterectomy, 1.2%-4.6% across patient venous thromboembolism risk groups). Venous thromboembolism risk was <0.5% for 30 (60%) of the procedures; 0.5% to 1.0% for 10 (20%) procedures; and >1.0% for 10 (20%) procedures. The risk for bleeding the require reintervention varied from <0.1% (transvaginal oocyte retrieval) to 4.0% (open myomectomy). The bleeding requiring reintervention risk was <0.5% in 17 (49%) procedures, 0.5% to 1.0% for 12 (34%) procedures, and >1.0% in 6 (17%) procedures. CONCLUSION: The risk for venous thromboembolism in gynecologic noncancer surgery varied between procedures and patients. Venous thromboembolism risks exceeded the bleeding risks only among selected patients and procedures. Although most of the evidence is of low certainty, the results nevertheless provide a compelling rationale for restricting pharmacologic thromboprophylaxis to a minority of patients who undergo gynecologic noncancer procedures.

Systematic Reviews and Meta-analyses of the Procedure-specific Risks of Thrombosis and Bleeding in General Abdominal, Colorectal, Upper Gastrointestinal, and Hepatopancreatobiliary Surgery.

OBJECTIVE: To provide procedure-specific estimates of symptomatic venous thromboembolism (VTE) and major bleeding after abdominal surgery. BACKGROUND: The use of pharmacological thromboprophylaxis represents a trade-off that depends on VTE and bleeding risks that vary between procedures; their magnitude remains uncertain. METHODS: We identified observational studies reporting procedure-specific risks of symptomatic VTE or major bleeding after abdominal surgery, adjusted the reported estimates for thromboprophylaxis and length of follow-up, and estimated cumulative incidence at 4 weeks postsurgery, stratified by VTE risk groups, and rated evidence certainty. RESULTS: After eligibility screening, 285 studies (8,048,635 patients) reporting on 40 general abdominal, 36 colorectal, 15 upper gastrointestinal, and 24 hepatopancreatobiliary surgery procedures proved eligible. Evidence certainty proved generally moderate or low for VTE and low or very low for bleeding requiring reintervention. The risk of VTE varied substantially among procedures: in general abdominal surgery from a median of <0.1% in laparoscopic cholecystectomy to a median of 3.7% in open small bowel resection, in colorectal from 0.3% in minimally invasive sigmoid colectomy to 10.0% in emergency open total proctocolectomy, and in upper gastrointestinal/hepatopancreatobiliary from 0.2% in laparoscopic sleeve gastrectomy to 6.8% in open distal pancreatectomy for cancer. CONCLUSIONS: VTE thromboprophylaxis provides net benefit through VTE reduction with a small increase in bleeding in some procedures (eg, open colectomy and open pancreaticoduodenectomy), whereas the opposite is true in others (eg, laparoscopic cholecystectomy and elective groin hernia repairs). In many procedures, thromboembolism and bleeding risks are similar, and decisions depend on individual risk prediction and values and preferences regarding VTE and bleeding.

MMPatho: Leveraging Multilevel Consensus and Evolutionary Information for Enhanced Missense Mutation Pathogenic Prediction.

Understanding the pathogenicity of missense mutation (MM) is essential for shed light on genetic diseases, gene functions, and individual variations. In this study, we propose a novel computational approach, called MMPatho, for enhancing missense mutation pathogenic prediction. First, we established a large-scale nonredundant MM benchmark data set based on the entire Ensembl database, complemented by a focused blind test set specifically for pathogenic GOF/LOF MM. Based on this data set, for each mutation, we utilized Ensembl VEP v104 and dbNSFP v4.1a to extract variant-level, amino acid-level, individuals' outputs, and genome-level features. Additionally, protein sequences were generated using ENSP identifiers with the Ensembl API, and then encoded. The mutant sites' ESM-1b and ProtTrans-T5 embeddings were subsequently extracted. Then, our model group (MMPatho) was developed by leveraging upon these efforts, which comprised ConsMM and EvoIndMM. To be specific, ConsMM employs individuals' outputs and XGBoost with SHAP explanation analysis, while EvoIndMM investigates the potential enhancement of predictive capability by incorporating evolutionary information from ESM-1b and ProtT5-XL-U50, large protein language embeddings. Through rigorous comparative experiments, both ConsMM and EvoIndMM were capable of achieving remarkable AUROC (0.9836 and 0.9854) and AUPR (0.9852 and 0.9902) values on the blind test set devoid of overlapping variations and proteins from the training data, thus highlighting the superiority of our computational approach in the prediction of MM pathogenicity. Our Web server, available at http://csbio.njust.edu.cn/bioinf/mmpatho/, allows researchers to predict the pathogenicity (alongside the reliability index score) of MMs using the ConsMM and EvoIndMM models and provides extensive annotations for user input. Additionally, the newly constructed benchmark data set and blind test set can be accessed via the data page of our web server.

Investigation on the Ductility Capacity of Concrete Columns with High Strength Steel Reinforcement under Eccentric Loading.

Ductility-based structural design is currently the mainstream method. In order to analyze the ductility performance of concrete columns with high-strength steel reinforcements under eccentric compression, corresponding experimental studies have been performed. Numerical models were established, and their reliability was verified. Based on the numerical models, the parameter analysis was carried out, where eccentricity, concrete strength, and reinforcement ratio were considered to systematically discuss the ductility of the concrete column section with high-strength steel reinforcement. The results show that the ductility of the section under eccentric compression increases with the strength of the concrete and eccentricity, and decreases with the reinforcement ratio. Finally, a simplified calculation formula capable of quantitatively evaluating the section ductility was proposed.

Prediction of Multiple Types of RNA Modifications via Biological Language Model.

It has been demonstrated that RNA modifications play essential roles in multiple biological processes. Accurate identification of RNA modifications in the transcriptome is critical for providing insights into the biological functions and mechanisms. Many tools have been developed for predicting RNA modifications at single-base resolution, which employ conventional feature engineering methods that focus on feature design and feature selection processes that require extensive biological expertise and may introduce redundant information. With the rapid development of artificial intelligence technologies, end-to-end methods are favorably received by researchers. Nevertheless, each well-trained model is only suitable for a specific RNA methylation modification type for nearly all of these approaches. In this study, we present MRM-BERT by feeding task-specific sequences into the powerful BERT (Bidirectional Encoder Representations from Transformers) model and implementing fine-tuning, which exhibits competitive performance to the state-of-the-art methods. MRM-BERT avoids repeated de novo training of the model and can predict multiple RNA modifications such as pseudouridine, m6A, m5C, and m1A in Mus musculus, Arabidopsis thaliana, and Saccharomyces cerevisiae. In addition, we analyse the attention heads to provide high attention regions for the prediction, and conduct saturated in silico mutagenesis of the input sequences to discover potential changes of RNA modifications, which can better assist researchers in their follow-up research.

[Characteristics of Phytoplankton Communities and Key Impact Factors in Three Types of Lakes in Wuhan].

To reveal the characteristics and key impact factors of phytoplankton communities in different types of lakes, sampling surveys for phytoplankton and water quality parameters were conducted at 174 sampling sites in a total of 24 lakes covering urban, countryside, and ecological conservation areas of Wuhan in spring, summer, autumn, and winter 2018. The results showed that a total of 365 species of phytoplankton from nine phyla and 159 genera were identified in the three types of lakes. The main species were green algae, cyanobacteria, and diatoms, accounting for 55.34%, 15.89%, and 15.07% of the total number of species, respectively. The phytoplankton cell density varied from 3.60×106-421.99×106 cell·L-1, chlorophyll-a content varied from 15.60-240.50 µg·L-1, biomass varied from 27.71-379.79 mg·L-1, and the Shannon-Wiener diversity index varied from 0.29-2.86. In the three lake types, cell density, Chla, and biomass were lower in EL and UL, whereas the opposite was true for the Shannon-Wiener diversity index. NMDS and ANOSIM analysis showed differences in phytoplankton community structure (Stress=0.13, R=0.048, P=0.2298). In addition, the phytoplankton community structure of the three lake types had significant seasonal characteristics, with chlorophyll-a content and biomass being significantly higher in summer than in winter (P<0.05). Spearman correlation analysis showed that phytoplankton biomass decreased with increasing N:P in UL and CL, whereas the opposite was true for EL. Redundancy analysis (RDA) showed that WT, pH, NO3-, EC, and N:P were the key factors that significantly affected the variability in phytoplankton community structure in the three types of lakes in Wuhan (P<0.05).

VPatho: a deep learning-based two-stage approach for accurate prediction of gain-of-function and loss-of-function variants.

Determining the pathogenicity and functional impact (i.e. gain-of-function; GOF or loss-of-function; LOF) of a variant is vital for unraveling the genetic level mechanisms of human diseases. To provide a 'one-stop' framework for the accurate identification of pathogenicity and functional impact of variants, we developed a two-stage deep-learning-based computational solution, termed VPatho, which was trained using a total of 9619 pathogenic GOF/LOF and 138 026 neutral variants curated from various databases. A total number of 138 variant-level, 262 protein-level and 103 genome-level features were extracted for constructing the models of VPatho. The development of VPatho consists of two stages: (i) a random under-sampling multi-scale residual neural network (ResNet) with a newly defined weighted-loss function (RUS-Wg-MSResNet) was proposed to predict variants' pathogenicity on the gnomAD_NV + GOF/LOF dataset; and (ii) an XGBOD model was constructed to predict the functional impact of the given variants. Benchmarking experiments demonstrated that RUS-Wg-MSResNet achieved the highest prediction performance with the weights calculated based on the ratios of neutral versus pathogenic variants. Independent tests showed that both RUS-Wg-MSResNet and XGBOD achieved outstanding performance. Moreover, assessed using variants from the CAGI6 competition, RUS-Wg-MSResNet achieved superior performance compared to state-of-the-art predictors. The fine-trained XGBOD models were further used to blind test the whole LOF data downloaded from gnomAD and accordingly, we identified 31 nonLOF variants that were previously labeled as LOF/uncertain variants. As an implementation of the developed approach, a webserver of VPatho is made publicly available at http://csbio.njust.edu.cn/bioinf/vpatho/ to facilitate community-wide efforts for profiling and prioritizing the query variants with respect to their pathogenicity and functional impact.