A merged molecular representation deep learning method for blood-brain barrier permeability prediction.

The ability of a compound to permeate across the blood-brain barrier (BBB) is a significant factor for central nervous system drug development. Thus, for speeding up the drug discovery process, it is crucial to perform high-throughput screenings to predict the BBB permeability of the candidate compounds. Although experimental methods are capable of determining BBB permeability, they are still cost-ineffective and time-consuming. To complement the shortcomings of existing methods, we present a deep learning-based multi-model framework model, called Deep-B3, to predict the BBB permeability of candidate compounds. In Deep-B3, the samples are encoded in three kinds of features, namely molecular descriptors and fingerprints, molecular graph and simplified molecular input line entry system (SMILES) text notation. The pre-trained models were built to extract latent features from the molecular graph and SMILES. These features depicted the compounds in terms of tabular data, image and text, respectively. The validation results yielded from the independent dataset demonstrated that the performance of Deep-B3 is superior to that of the state-of-the-art models. Hence, Deep-B3 holds the potential to become a useful tool for drug development. A freely available online web-server for Deep-B3 was established at http://cbcb.cdutcm.edu.cn/deepb3/, and the source code and dataset of Deep-B3 are available at https://github.com/GreatChenLab/Deep-B3.

DeepLncPro: an interpretable convolutional neural network model for identifying long non-coding RNA promoters.

Long non-coding RNA (lncRNA) plays important roles in a series of biological processes. The transcription of lncRNA is regulated by its promoter. Hence, accurate identification of lncRNA promoter will be helpful to understand its regulatory mechanisms. Since experimental techniques remain time consuming for gnome-wide promoter identification, developing computational tools to identify promoters are necessary. However, only few computational methods have been proposed for lncRNA promoter prediction and their performances still have room to be improved. In the present work, a convolutional neural network based model, called DeepLncPro, was proposed to identify lncRNA promoters in human and mouse. Comparative results demonstrated that DeepLncPro was superior to both state-of-the-art machine learning methods and existing models for identifying lncRNA promoters. Furthermore, DeepLncPro has the ability to extract and analyze transcription factor binding motifs from lncRNAs, which made it become an interpretable model. These results indicate that the DeepLncPro can server as a powerful tool for identifying lncRNA promoters. An open-source tool for DeepLncPro was provided at https://github.com/zhangtian-yang/DeepLncPro.

The evolution of N6-methyladenosine regulators in plants.

As a reversible modification, N6-methyladenosine (m6A) plays key roles in series of biological processes. Although it has been found that m6A modification is regulated by writers, erasers and readers, their evolutionary processes are still not clearly and systematically described. In the present work, we identified 1592 m6A modification regulators from 65 representative plant species and performed the phylogenetic relationships, sequence structure, selection pressure, and codon usage analysis across species. The regulators from different species or subfamilies were distinguishable based on the phylogenetic trees. Although the gene structure was structurally and functionally conserved for each kind of regulators, the unique exon/intron structures and motif organizations were observed among different families. The selection pressure analysis demonstrated that the regulators experienced purifying selection. Interestingly, the selection pressure for the regulators in higher plants was more relaxed, indicating that they might have acquired new functions during evolution. In addition, the different codon usage preferences were observed for the different kinds of m6A modification regulators. These results will not only facilitate our understanding of the evolution of m6A regulators, but also shed light on how the evolutionary differences affect their functional divergence.

Brassica carinata genome characterization clarifies U's triangle model of evolution and polyploidy in Brassica.

Ethiopian mustard (Brassica carinata) in the Brassicaceae family possesses many excellent agronomic traits. Here, the high-quality genome sequence of B. carinata is reported. Characterization revealed a genome anchored to 17 chromosomes with a total length of 1.087 Gb and an N50 scaffold length of 60 Mb. Repetitive sequences account for approximately 634 Mb or 58.34% of the B. carinata genome. Notably, 51.91% of 97,149 genes are confined to the terminal 20% of chromosomes as a result of the expansion of repeats in pericentromeric regions. Brassica carinata shares one whole-genome triplication event with the five other species in U's triangle, a classic model of evolution and polyploidy in Brassica. Brassica carinata was deduced to have formed ∼0.047 Mya, which is slightly earlier than B. napus but later than B. juncea. Our analysis indicated that the relationship between the two subgenomes (BcaB and BcaC) is greater than that between other two tetraploid subgenomes (BjuB and BnaC) and their respective diploid parents. RNA-seq datasets and comparative genomic analysis were used to identify several key genes in pathways regulating disease resistance and glucosinolate metabolism. Further analyses revealed that genome triplication and tandem duplication played important roles in the expansion of those genes in Brassica species. With the genome sequencing of B. carinata completed, the genomes of all six Brassica species in U's triangle are now resolved. The data obtained from genome sequencing, transcriptome analysis, and comparative genomic efforts in this study provide valuable insights into the genome evolution of the six Brassica species in U's triangle.

mRNALocater: Enhance the prediction accuracy of eukaryotic mRNA subcellular localization by using model fusion strategy.

The functions of mRNAs are closely correlated with their locations in cells. Knowledge about the subcellular locations of mRNA is helpful to understand their biological functions. In recent years, it has become a hot topic to develop effective computational models to predict eukaryotic mRNA subcellular localizations. However, existing state-of-the-art models still have certain deficiencies in terms of prediction accuracy and generalization ability. Therefore, it is urgent to develop novel methods to accurately predict mRNA subcellular localizations. In this study, a novel method called mRNALocater was proposed to detect the subcellular localization of eukaryotic mRNA by adopting the model fusion strategy. To fully extract information from mRNA sequences, the electron-ion interaction pseudopotential and pseudo k-tuple nucleotide composition were used to encode the sequences. Moreover, the correlation coefficient filtering algorithm and feature forward search technology were used to mine hidden feature information, which guarantees that mRNALocater can be more effectively applied to new sequences. The results based on the independent dataset tests demonstrate that mRNALocater yields promising performances for predicting eukaryotic mRNA subcellular localizations and is a powerful tool in practical applications. A freely available online web server for mRNALocater has been established at http://bio-bigdata.cn/mRNALocater.

The celery genome sequence reveals sequential paleo-polyploidizations, karyotype evolution and resistance gene reduction in apiales.

Celery (Apium graveolens L. 2n = 2x = 22), a member of the Apiaceae family, is among the most important and globally grown vegetables. Here, we report a high-quality genome sequence assembly, anchored to 11 chromosomes, with total length of 3.33 Gb and N50 scaffold length of 289.78 Mb. Most (92.91%) of the genome is composed of repetitive sequences, with 62.12% of 31 326 annotated genes confined to the terminal 20% of chromosomes. Simultaneous bursts of shared long-terminal repeats (LTRs) in different Apiaceae plants suggest inter-specific exchanges. Two ancestral polyploidizations were inferred, one shared by Apiales taxa and the other confined to Apiaceae. We reconstructed 8 Apiales proto-chromosomes, inferring their evolutionary trajectories from the eudicot common ancestor to extant plants. Transcriptome sequencing in three tissues (roots, leaves and petioles), and varieties with different-coloured petioles, revealed 4 and 2 key genes in pathways regulating anthocyanin and coumarin biosynthesis, respectively. A remarkable paucity of NBS disease-resistant genes in celery (62) and other Apiales was explained by extensive loss and limited production of these genes during the last ~10 million years, raising questions about their biotic defence mechanisms and motivating research into effects of chemicals, for example coumarins, that give off distinctive odours. Celery genome sequencing and annotation facilitates further research into important gene functions and breeding, and comparative genomic analyses in Apiales.

Deciphering the high-quality genome sequence of coriander that causes controversial feelings.

Coriander (Coriandrum sativum L. 2n = 2x = 22), a plant from the Apiaceae family, also called cilantro or Chinese parsley, is a globally important crop used as vegetable, spice, fragrance and traditional medicine. Here, we report a high-quality assembly and analysis of its genome sequence, anchored to 11 chromosomes, with total length of 2118.68 Mb and N50 scaffold length of 160.99 Mb. We found that two whole-genome duplication events, respectively, dated to ~45-52 and ~54-61 million years ago, were shared by the Apiaceae family after their split from lettuce. Unbalanced gene loss and expression are observed between duplicated copies produced by these two events. Gene retention, expression, metabolomics and comparative genomic analyses of terpene synthase (TPS) gene family, involved in terpenoid biosynthesis pathway contributing to coriander's special flavour, revealed that tandem duplication contributed to coriander TPS gene family expansion, especially compared to their carrot counterparts. Notably, a TPS gene highly expressed in all 4 tissues and 3 development stages studied is likely a major-effect gene encoding linalool synthase and myrcene synthase. The present genome sequencing, transcriptome, metabolome and comparative genomic efforts provide valuable insights into the genome evolution and spice trait biology of Apiaceae and other related plants, and facilitated further research into important gene functions and crop improvement.

i6mA-Pred: identifying DNA N6-methyladenine sites in the rice genome.

MOTIVATION: DNA N6-methyladenine (6mA) is associated with a wide range of biological processes. Since the distribution of 6mA site in the genome is non-random, accurate identification of 6mA sites is crucial for understanding its biological functions. Although experimental methods have been proposed for this regard, they are still cost-ineffective for detecting 6mA site in genome-wide scope. Therefore, it is desirable to develop computational methods to facilitate the identification of 6mA site. RESULTS: In this study, a computational method called i6mA-Pred was developed to identify 6mA sites in the rice genome, in which the optimal nucleotide chemical properties obtained by the using feature selection technique were used to encode the DNA sequences. It was observed that the i6mA-Pred yielded an accuracy of 83.13% in the jackknife test. Meanwhile, the performance of i6mA-Pred was also superior to other methods. AVAILABILITY AND IMPLEMENTATION: A user-friendly web-server, i6mA-Pred is freely accessible at http://lin-group.cn/server/i6mA-Pred.

MimoDB 2.0: a mimotope database and beyond.

Mimotopes are peptides with affinities to given targets. They are readily obtained through biopanning against combinatorial peptide libraries constructed by phage display and other display technologies such as mRNA display, ribosome display, bacterial display and yeast display. Mimotopes have been used to infer the protein interaction sites and networks; they are also ideal candidates for developing new diagnostics, therapeutics and vaccines. However, such valuable peptides are not collected in the central data resources such as UniProt and NCBI GenPept due to their 'unnatural' short sequences. The MimoDB database is an information portal to biopanning results of random libraries. In version 2.0, it has 15,633 peptides collected from 849 papers and grouped into 1818 sets. Besides the core data on panning experiments and their results, broad background information on target, template, library and structure is included. An accompanied benchmark has also been compiled for bioinformaticians to develop and evaluate their new models, algorithms and programs. In addition, the MimoDB database provides tools for simple and advanced searches, structure visualization, BLAST and alignment view on the fly. The experimental biologists can easily use the database as a virtual control to exclude possible target-unrelated peptides. The MimoDB database is freely available at http://immunet.cn/mimodb.

PHGD: An integrative and user-friendly database for plant hormone-related genes.

Plant Hormone Gene Database (PHGD) database platform construction pipeline. First, we collected all reported hormone-related genes in the model plant Arabidopsis thaliana, and combined with the existing experimental background, mapped the hormone-gene interaction network to provide a blueprint. Next, we collected 469 high-quality plant genomes. Then, bioinformatics was used to identify hormone-related genes in these plants. Finally, these genetic data were programmed to be stored in a database and a platform website PHGD was built. PHGD was divided into eight modules, namely Home, Browse, Search, Resources, Download, Tools, Help, and Contact. We provided data resources and platform services to facilitate the study of plant hormones.

Flowering genes identification, network analysis, and database construction for 837 plants.

Flowering is one of the most important biological phenomena in the plant kingdom, which not only has important ecological significance, but also has substantial horticultural ornamental value. In this study, we undertook an exhaustive review of the advancements in our understanding of plant flowering genes. We delved into the identification and conducted comparative analyses of flowering genes across virtually all sequenced angiosperm plant genomes. Furthermore, we established an extensive angiosperm flowering atlas, encompassing a staggering 183 720 genes across eight pathways, along with 10 155 ABCDE mode genes, which play a pivotal role in plant flowering regulation. Through the examination of expression patterns, we unveiled the specificities of these flowering genes. An interaction network between flowering genes of the ABCDE model and their corresponding upstream genes offered a blueprint for comprehending their regulatory mechanisms. Moreover, we predicted the miRNA and target genes linked to the flowering processes of each species. To culminate our efforts, we have built a user-friendly web interface, named the Plant Flowering-time Gene Database (PFGD), accessible at http://pfgd.bio2db.com/. We firmly believe that this database will serve as a cornerstone in the global research community, facilitating the in-depth exploration of flowering genes in the plant kingdom. In summation, this pioneering endeavor represents the first comprehensive collection and comparative analysis of flowering genes in plants, offering valuable resources for the study of plant flowering genetics.

HODD: A Manually Curated Database of Human Ophthalmic Diseases with Symptom Characteristics and Genetic Variants Towards Facilitating Quick and Definite Diagnosis.

Ophthalmic diseases are disorders that affect the eyes. Hundreds of causal genes and biological pathways have been reported to be closely correlated with ophthalmic diseases. However, these information are scattered across various resources, which has hindered a thorough and deep understanding of ophthalmic diseases. In the present work, we proposed the Human Ophthalmic Diseases Database (HODD), which currently deposits 730 ophthalmic diseases and 653 related genes and is available at http://bio-bigdata.cn/HODD/ . The disease-related information and genes related to ophthalmic diseases were collected from the several well-known databases. To comprehensively understand the ophthalmic diseases, the basic information was provided for each disease, including disease description, related genes, gene location, ocular and extraocular effect of the disease, protein-protein interaction and disease-associated pathways. All these data were reorganized and made accessible through multiple entrances. We hope that HODD will facilitate studies on ophthalmic diseases. The workflow for the construction of the HODD (Human Ophthalmic Diseases Database, http://bio-bigdata.cn/HODD/ ) database.

RNAME: A comprehensive database of RNA modification enzymes.

The dynamic RNA modifications were orchestrated by a series of enzymes, namely "writer", "reader" and "eraser", which can install, recognize and remove the modifications, respectively. However, only a very small number of experimentally validated RNA modification enzymes have been identified and reported. Therefore, there is an urgent need to develop a database to deposit RNA modification enzymes. In the present work, we developed the RNAME database (https://chenweilab.cn/rname/) to provide a comprehensive resource for RNA modification enzymes. The current version of RNAME deposits more than 21,000 manually curated RNA modification enzymes, which are from 456 species and covers the 7 common kinds of RNA modifications (i.e., adenosine to inosine, N1-methyladenosine, N6-methyladenosine, 5-methylcytidine, N7-methylguanosine, mRNA cap modification, and pseudouridine). The 3D structures, domains, subcellular locations, and biological functions of these enzymes were also integrated in RNAME. It is anticipated that RNAME will facilitate the researches on RNA modifications.

TVIR: a comprehensive vegetable information resource database for comparative and functional genomic studies.

Vegetables are an indispensable part of the daily diet of humans. Therefore, it is vital to systematically study the genomic data of vegetables and build a platform for data sharing and analysis. In this study, a comprehensive platform for vegetables with a user-friendly Web interface-The Vegetable Information Resource (TVIR, http://tvir.bio2db.com)-was built based on the genomes of 59 vegetables. TVIR database contains numerous important functional genes, including 5215 auxin genes, 2437 anthocyanin genes, 15 002 flowering genes, 79 830 resistance genes, and 2639 glucosinolate genes of 59 vegetables. In addition, 2597 N6-methyladenosine (m6A) genes were identified, including 513 writers, 1058 erasers, and 1026 readers. A total of 2 101 501 specific clustered regularly interspaced short palindromic repeat (CRISPR) guide sequences and 17 377 miRNAs were detected and deposited in TVIR database. Information on gene synteny, duplication, and orthologs is also provided for 59 vegetable species. TVIR database contains 2 346 850 gene annotations by the Swiss-Prot, TrEMBL, Gene Ontology (GO), Pfam, and Non-redundant (Nr) databases. Synteny, Primer Design, Blast, and JBrowse tools are provided to facilitate users in conducting comparative genomic analyses. This is the first large-scale collection of vegetable genomic data and bioinformatic analysis. All genome and gene sequences, annotations, and bioinformatic results can be easily downloaded from TVIR. Furthermore, transcriptome data of 98 vegetables have been collected and collated, and can be searched by species, tissues, or different growth stages. TVIR is expected to become a key hub for vegetable research globally. The database will be updated with newly assembled vegetable genomes and comparative genomic studies in the future.

ncPro-ML: An integrated computational tool for identifying non-coding RNA promoters in multiple species.

The promoter is located near the transcription start sites and regulates transcription initiation of the gene. Accurate identification of promoters is essential for understanding the mechanism of gene regulation. Since experimental methods are costly and ineffective, developing efficient and accurate computational tools to identify promoters are necessary. Although a series of methods have been proposed for identifying promoters, none of them is able to identify the promoters of non-coding RNA (ncRNA). In the present work, a new method called ncPro-ML was proposed to identify the promoter of ncRNA in Homo sapiens and Mus musculus, in which different kinds of sequence encoding schemes were used to convert DNA sequences into feature vectors. To test the length effect, for each species, datasets including sequences with different lengths were built. The results demonstrated that ncPro-ML achieved the best performance based on the dataset with the sequence length of 221 nucleotides for human and mouse. The performances of ncPro-ML were also satisfying from both independent dataset test and cross-species test. The results indicate that the proposed predictor can server as a powerful tool for the discovery of ncRNA promoters. In addition, a web-server for ncPro-ML was developed, which can be freely accessed at http://www.bio-bigdata.cn/ncPro-ML/.

iATP: A Sequence Based Method for Identifying Anti-tubercular Peptides.

BACKGROUND: Tuberculosis is one of the biggest threats to human health. Recent studies have demonstrated that anti-tubercular peptides are promising candidates for the discovery of new anti-tubercular drugs. Since experimental methods are still labor intensive, it is highly desirable to develop automatic computational methods to identify anti-tubercular peptides from the huge amount of natural and synthetic peptides. Hence, accurate and fast computational methods are highly needed. METHODS AND RESULTS: In this study, a support vector machine based method was proposed to identify anti-tubercular peptides, in which the peptides were encoded by using the optimal g-gap dipeptide compositions. Comparative results demonstrated that our method outperforms existing methods on the same benchmark dataset. For the convenience of scientific community, a freely accessible web-server was built, which is available at http://lin-group.cn/server/iATP. CONCLUSION: It is anticipated that the proposed method will become a useful tool for identifying anti-tubercular peptides.

Recent Advances of Computational Methods for Identifying Bacteriophage Virion Proteins.

Phage Virion Proteins (PVP) are essential materials of bacteriophage, which participate in a series of biological processes. Accurate identification of phage virion proteins is helpful to understand the mechanism of interaction between the phage and its host bacteria. Since experimental method is labor intensive and time-consuming, in the past few years, many computational approaches have been proposed to identify phage virion proteins. In order to facilitate researchers to select appropriate methods, it is necessary to give a comprehensive review and comparison on existing computational methods on identifying phage virion proteins. In this review, we summarized the existing computational methods for identifying phage virion proteins and also assessed their performances on an independent dataset. Finally, challenges and future perspectives for identifying phage virion proteins were presented. Taken together, we hope that this review could provide clues to researches on the study of phage virion proteins.

RNAWRE: a resource of writers, readers and erasers of RNA modifications.

RNA modifications are involved in various kinds of cellular biological processes. Accumulated evidences have demonstrated that the functions of RNA modifications are determined by the effectors that can catalyze, recognize and remove RNA modifications. They are called 'writers', 'readers' and 'erasers'. The identification of RNA modification effectors will be helpful for understanding the regulatory mechanisms and biological functions of RNA modifications. In this work, we developed a database called RNAWRE that specially deposits RNA modification effectors. The current version of RNAWRE stored 2045 manually curated writers, readers and erasers for the six major kinds of RNA modifications, namely Cap, m1A, m6A, m5C, ψ and Poly A. The main modules of RNAWRE not only allow browsing and downloading the RNA modification effectors but also support the BLAST search of the potential RNA modification effectors in other species. We hope that RNAWRE will be helpful for the researches on RNA modifications. Database URL: http://rnawre.bio2db.com.

Coriander Genomics Database: a genomic, transcriptomic, and metabolic database for coriander.

Coriander (Coriandrum sativum L.), also known as cilantro, is a globally important vegetable and spice crop. Its genome and that of carrot are models for studying the evolution of the Apiaceae family. Here, we developed the Coriander Genomics Database (CGDB, http://cgdb.bio2db.com/) to collect, store, and integrate the genomic, transcriptomic, metabolic, functional annotation, and repeat sequence data of coriander and carrot to serve as a central online platform for Apiaceae and other related plants. Using these data sets in the CGDB, we intriguingly found that seven transcription factor (TF) families showed significantly greater numbers of members in the coriander genome than in the carrot genome. The highest ratio of the numbers of MADS TFs between coriander and carrot reached 3.15, followed by those for tubby protein (TUB) and heat shock factors. As a demonstration of CGDB applications, we identified 17 TUB family genes and conducted systematic comparative and evolutionary analyses. RNA-seq data deposited in the CGDB also suggest dose compensation effects of gene expression in coriander. CGDB allows bulk downloading, significance searches, genome browser analyses, and BLAST searches for comparisons between coriander and other plants regarding genomics, gene families, gene collinearity, gene expression, and the metabolome. A detailed user manual and contact information are also available to provide support to the scientific research community and address scientific questions. CGDB will be continuously updated, and new data will be integrated for comparative and functional genomic analysis in Apiaceae and other related plants.

NIEluter: Predicting peptides eluted from HLA class I molecules.

The immune system has evolved to make a diverse repertoire of peptides processed from self and foreign proteomes, which are displayed in antigen-binding grooves of major histocompatibility complex (MHC) proteins at cell surface for surveillance by T cells. These antigenic peptides are termed Naturally Processed Peptides or Naturally Presented Peptides (NPPs), which play a major role in cell-mediated immunity and rational vaccine design. Therefore, it is intensely desirable to predict NPPs from a given protein antigen, or to foretell if an MHC-binding peptide can be eluted from a given MHC protein. In this paper, we describe NIEluter, an ensemble predictor based on support vector machine (SVM). It consists of a combination of five SVM models trained with position-specific amino acid composition, position-specific dipeptide composition, Hidden Markov Model, binary encoding, and BLOSUM62 feature. NIEluter can predict NPPs of length 8-11 from six HLA alleles (A0201, B0702, B3501, B4403, B5301, and B5701) at present. Evaluated with five-fold cross-validation and independent datasets if available, NIEluter shows good performance. It outperforms MHC-NP in 7 out of 24 types of situation and precedes NetMHC3.2 in most cases, indicating that it is a helpful complement to available tools. NIEluter has been implemented as a free web service, which can be accessed at http://immunet.cn/nie/cgi-bin/nieluter.pl.