Chuan Dong

Accurate prediction of human essential genes using only nucleotide composition and association information

Motivation: Previously constructed classifiers in predicting eukaryotic essential genes integrated a variety of features including experimental ones. If we can obtain satisfactory prediction using only nucleotide (sequence) information, it would be more promising. Three groups recently identified essential genes in human cancer cell lines using wet experiments and it provided wonderful opportunity to accomplish our idea. Here we improved the Z curve method into the λ‐interval form to denote nucleotide composition and association information and used it to construct the SVM classifying model. Results: Our model accurately predicted human gene essentiality with an AUC higher than 0.88 both for 5‐fold cross‐validation and jackknife tests. These results demonstrated that the essentiality of human genes could be reliably reflected by only sequence information. We re‐predicted the negative dataset by our Pheg server and 118 genes were additionally predicted as essential. Among them, 20 were found to be homologues in mouse essential genes, indicating that some of the 118 genes were indeed essential, however previous experiments overlooked them. As the first available server, Pheg could predict essentiality for anonymous gene sequences of human. It is also hoped the λ‐interval Z curve method could be effectively extended to classification issues of other DNA elements. Availability and Implementation: http://cefg.uestc.edu.cn/Pheg Contact: fbguo@uestc.edu.cn Supplementary information: Supplementary data are available at Bioinformatics online.

A novel proposal of a simplified bacterial gene set and the neo-construction of a general minimized metabolic network

A minimal gene set (MGS) is critical for the assembly of a minimal artificial cell. We have developed a proposal of simplifying bacterial gene set to approximate a bacterial MGS by the following procedure. First, we base our simplified bacterial gene set (SBGS) on experimentally determined essential genes to ensure that the genes included in the SBGS are critical. Second, we introduced a half-retaining strategy to extract persistent essential genes to ensure stability. Third, we constructed a viable metabolic network to supplement SBGS. The proposed SBGS includes 327 genes and required 431 reactions. This report describes an SBGS that preserves both self-replication and self-maintenance systems. In the minimized metabolic network, we identified five novel hub metabolites and confirmed 20 known hubs. Highly essential genes were found to distribute the connecting metabolites into more reactions. Based on our SBGS, we expanded the pool of targets for designing broad-spectrum antibacterial drugs to reduce pathogen resistance. We also suggested a rough semi-de novo strategy to synthesize an artificial cell, with potential applications in industry.

IMGMD: A platform for the integration and standardisation of In silico Microbial Genome-scale Metabolic Models

Genome-scale metabolic models (GSMMs) constitute a platform that combines genome sequences and detailed biochemical information to quantify microbial physiology at the system level. To improve the unity, integrity, correctness, and format of data in published GSMMs, a consensus IMGMD database was built in the LAMP (Linux + Apache + MySQL + PHP) system by integrating and standardizing 328 GSMMs constructed for 139 microorganisms. The IMGMD database can help microbial researchers download manually curated GSMMs, rapidly reconstruct standard GSMMs, design pathways, and identify metabolic targets for strategies on strain improvement. Moreover, the IMGMD database facilitates the integration of wet-lab and in silico data to gain an additional insight into microbial physiology. The IMGMD database is freely available, without any registration requirements, at http://imgmd.jiangnan.edu.cn/database.

Identification and analysis of genomic islands in Burkholderia cenocepacia AU 1054 with emphasis on pathogenicity islands

BackgroundGenomic islands (GIs) are genomic regions that reveal evidence of horizontal DNA transfer. They can code for many functions and may augment a bacterium’s adaptation to its host or environment. GIs have been identified in strain J2315 of Burkholderia cenocepacia, whereas in strain AU 1054 there has been no published works on such regions according to our text mining and keyword search in Medline.ResultsIn this study, we identified 21 GIs in AU 1054 by combining two computational tools. Feature analyses suggested that the predictions are highly reliable and hence illustrated the advantage of joint predictions by two independent methods. Based on putative virulence factors, four GIs were further identified as pathogenicity islands (PAIs). Through experiments of gene deletion mutants in live bacteria, two putative PAIs were confirmed, and the virulence factors involved were identified as lipA and copR. The importance of the genes lipA (from PAI 1) and copR (from PAI 2) for bacterial invasion and replication indicates that they are required for the invasive properties of B. cenocepacia and may function as virulence determinants for bacterial pathogenesis and host infection.ConclusionsThis approach of in silico prediction of GIs and subsequent identification of potential virulence factors in the putative island regions with final validation using wet experiments could be used as an effective strategy to rapidly discover novel virulence factors in other bacterial species and strains.

An Approach for Predicting Essential Genes Using Multiple Homology Mapping and Machine Learning Algorithms

Investigation of essential genes is significant to comprehend the minimal gene sets of cell and discover potential drug targets. In this study, a novel approach based on multiple homology mapping and machine learning method was introduced to predict essential genes. We focused on 25 bacteria which have characterized essential genes. The predictions yielded the highest area under receiver operating characteristic (ROC) curve (AUC) of 0.9716 through tenfold cross-validation test. Proper features were utilized to construct models to make predictions in distantly related bacteria. The accuracy of predictions was evaluated via the consistency of predictions and known essential genes of target species. The highest AUC of 0.9552 and average AUC of 0.8314 were achieved when making predictions across organisms. An independent dataset from Synechococcus elongatus, which was released recently, was obtained for further assessment of the performance of our model. The AUC score of predictions is 0.7855, which is higher than other methods. This research presents that features obtained by homology mapping uniquely can achieve quite great or even better results than those integrated features. Meanwhile, the work indicates that machine learning-based method can assign more efficient weight coefficients than using empirical formula based on biological knowledge.

Anti-CRISPRdb: a comprehensive online resource for anti-CRISPR proteins

Abstract CRISPR-Cas is a tool that is widely used for gene editing. However, unexpected off-target effects may occur as a result of long-term nuclease activity. Anti-CRISPR proteins, which are powerful molecules that inhibit the CRISPR–Cas system, may have the potential to promote better utilization of the CRISPR-Cas system in gene editing, especially for gene therapy. Additionally, more in-depth research on these proteins would help researchers to better understand the co-evolution of bacteria and phages. Therefore, it is necessary to collect and integrate data on various types of anti-CRISPRs. Herein, data on these proteins were manually gathered through data screening of the literatures. Then, the first online resource, anti-CRISPRdb, was constructed for effectively organizing these proteins. It contains the available protein sequences, DNA sequences, coding regions, source organisms, taxonomy, virulence, protein interactors and their corresponding three-dimensional structures. Users can access our database at http://cefg.uestc.edu.cn/anti-CRISPRdb/ without registration. We believe that the anti-CRISPRdb can be used as a resource to facilitate research on anti-CRISPR proteins and in related fields.

SSER: Species specific essential reactions database

BackgroundEssential reactions are vital components of cellular networks. They are the foundations of synthetic biology and are potential candidate targets for antimetabolic drug design. Especially if a single reaction is catalyzed by multiple enzymes, then inhibiting the reaction would be a better option than targeting the enzymes or the corresponding enzyme-encoding gene. The existing databases such as BRENDA, BiGG, KEGG, Bio-models, Biosilico, and many others offer useful and comprehensive information on biochemical reactions. But none of these databases especially focus on essential reactions. Therefore, building a centralized repository for this class of reactions would be of great value.DescriptionHere, we present a species-specific essential reactions database (SSER). The current version comprises essential biochemical and transport reactions of twenty-six organisms which are identified via flux balance analysis (FBA) combined with manual curation on experimentally validated metabolic network models. Quantitative data on the number of essential reactions, number of the essential reactions associated with their respective enzyme-encoding genes and shared essential reactions across organisms are the main contents of the database.ConclusionSSER would be a prime source to obtain essential reactions data and related gene and metabolite information and it can significantly facilitate the metabolic network models reconstruction and analysis, and drug target discovery studies. Users can browse, search, compare and download the essential reactions of organisms of their interest through the website http://cefg.uestc.edu.cn/sser.

Metabolic pathway databases and model repositories

BackgroundThe number of biological Knowledge bases/databases storing metabolic pathway information and models has been growing rapidly. These resources are diverse in the type of information/data, the analytical tools, and objectives. Here we present a review of the most popular metabolic pathway databases and model repositories, focusing on their scope, content including reactions, enzymes, compounds, and genes, and applicability. The review aims to help researchers choose a suitable database or model repository according to the information and data required, by providing an insight look of each pathway resource.ResultsFour pathways databases and three model repositories were selected on the basis of popularity and diversity. Our review showed that the pathway resources vary in many aspects, such as their scope, content, access to data and the tools. In addition, inconsistencies have been observed in nomenclature and representation of database entities. The three model repositories reviewed do not offer a brief description of the models’ characteristics such as simulation conditions.ConclusionsThe inconsistencies among the databases in representing their contents may hamper the maximal use of the knowledge accumulated in these databases in particular and the area of systems biology at large. Therefore, it is strongly recommended that the database creators and the metabolic network models developers should follow international standards for the nomenclature of reactions and metabolites. Besides, computationally generated models that could be obtained from model repositories should be utilized with manual curations as they lack some important components that are necessary for full functionality of the models.