Prediction Of Essential Genes Research Articles

Genome-scale metabolic models highlight stage-specific differences in essential metabolic pathways in Trypanosoma cruzi.

Chagas disease is a neglected tropical disease and a leading cause of heart failure in Latin America caused by a protozoan called Trypanosoma cruzi. This parasite presents a complex multi-stage life cycle. Anti-Chagas drugs currently available are limited to benznidazole and nifurtimox, both with severe side effects. Thus, there is a need for alternative and more efficient drugs. Genome-scale metabolic models (GEMs) can accurately predict metabolic capabilities and aid in drug discovery in metabolic genes. This work developed an extended GEM, hereafter referred to as iIS312, of the published and validated T. cruzi core metabolism model. From iIS312, we then built three stage-specific models through transcriptomics data integration, and showed that epimastigotes present the most active metabolism among the stages (see S1-S4 GEMs). Stage-specific models predicted significant metabolic differences among stages, including variations in flux distribution in core metabolism. Moreover, the gene essentiality predictions suggest potential drug targets, among which some have been previously proven lethal, including glutamate dehydrogenase, glucokinase and hexokinase. To validate the models, we measured the activity of enzymes in the core metabolism of the parasite at different stages, and showed the results were consistent with model predictions. Our results represent a potential step forward towards the improvement of Chagas disease treatment. To our knowledge, these stage-specific models are the first GEMs built for the stages Amastigote and Trypomastigote. This work is also the first to present an in silico GEM comparison among different stages in the T. cruzi life cycle.

Prediction of Essential Genes in Comparison States Using Machine Learning.

Identifying essential genes in comparison states (EGS) is vital to understanding cell differentiation, performing drug discovery, and identifying disease causes. Here, we present a machine learning method termed Prediction of Essential Genes in Comparison States (PreEGS). To capture the alteration of the network in comparison states, PreEGS extracts topological and gene expression features of each gene in a five-dimensional vector. PreEGS also recruits a positive sample expansion method to address the problem of unbalanced positive and negative samples, which is often encountered in practical applications. Different classifiers are applied to the simulated datasets, and the PreEGS based on the random forests model (PreEGSRF) was chosen for optimal performance. PreEGSRF was then compared with six other methods, including three machine learning methods, to predict EGS in a specific state. On real datasets with four gene regulatory networks, PreEGSRF predicted five essential genes related to leukemia and five enriched KEGG pathways. Four of the predicted essential genes and all predicted pathways were consistent with previous studies and highly correlated with leukemia. With high prediction accuracy and generalization ability, PreEGSRF is broadly applicable for the discovery of disease-causing genes, driver genes for cell fate decisions, and complex biomarkers of biological systems.

DeepHE: Accurately predicting human essential genes based on deep learning.

Accurately predicting essential genes using computational methods can greatly reduce the effort in finding them via wet experiments at both time and resource scales, and further accelerate the process of drug discovery. Several computational methods have been proposed for predicting essential genes in model organisms by integrating multiple biological data sources either via centrality measures or machine learning based methods. However, the methods aiming to predict human essential genes are still limited and the performance still need improve. In addition, most of the machine learning based essential gene prediction methods are lack of skills to handle the imbalanced learning issue inherent in the essential gene prediction problem, which might be one factor affecting their performance. We propose a deep learning based method, DeepHE, to predict human essential genes by integrating features derived from sequence data and protein-protein interaction (PPI) network. A deep learning based network embedding method is utilized to automatically learn features from PPI network. In addition, 89 sequence features were derived from DNA sequence and protein sequence for each gene. These two types of features are integrated to train a multilayer neural network. A cost-sensitive technique is used to address the imbalanced learning problem when training the deep neural network. The experimental results for predicting human essential genes show that our proposed method, DeepHE, can accurately predict human gene essentiality with an average performance of AUC higher than 94%, the area under precision-recall curve (AP) higher than 90%, and the accuracy higher than 90%. We also compare DeepHE with several widely used traditional machine learning models (SVM, Naïve Bayes, Random Forest, and Adaboost) using the same features and utilizing the same cost-sensitive technique to against the imbalanced learning issue. The experimental results show that DeepHE significantly outperforms the compared machine learning models. We have demonstrated that human essential genes can be accurately predicted by designing effective machine learning algorithm and integrating representative features captured from available biological data. The proposed deep learning framework is effective for such task.

Genome-Scale Metabolic Model of the Human Pathogen Candida albicans: A Promising Platform for Drug Target Prediction.

Candida albicans is one of the most impactful fungal pathogens and the most common cause of invasive candidiasis, which is associated with very high mortality rates. With the rise in the frequency of multidrug-resistant clinical isolates, the identification of new drug targets and new drugs is crucial in overcoming the increase in therapeutic failure. In this study, the first validated genome-scale metabolic model for Candida albicans, iRV781, is presented. The model consists of 1221 reactions, 926 metabolites, 781 genes, and four compartments. This model was reconstructed using the open-source software tool merlin 4.0.2. It is provided in the well-established systems biology markup language (SBML) format, thus, being usable in most metabolic engineering platforms, such as OptFlux or COBRA. The model was validated, proving accurate when predicting the capability of utilizing different carbon and nitrogen sources when compared to experimental data. Finally, this genome-scale metabolic reconstruction was tested as a platform for the identification of drug targets, through the comparison between known drug targets and the prediction of gene essentiality in conditions mimicking the human host. Altogether, this model provides a promising platform for global elucidation of the metabolic potential of C. albicans, possibly guiding the identification of new drug targets to tackle human candidiasis.

DeeplyEssential: a deep neural network for predicting essential genes in microbes

BackgroundEssential genes are those genes that are critical for the survival of an organism. The prediction of essential genes in bacteria can provide targets for the design of novel antibiotic compounds or antimicrobial strategies.ResultsWe propose a deep neural network for predicting essential genes in microbes. Our architecture called DeeplyEssential makes minimal assumptions about the input data (i.e., it only uses gene primary sequence and the corresponding protein sequence) to carry out the prediction thus maximizing its practical application compared to existing predictors that require structural or topological features which might not be readily available. We also expose and study a hidden performance bias that effected previous classifiers. Extensive results show that DeeplyEssential outperform existing classifiers that either employ down-sampling to balance the training set or use clustering to exclude multiple copies of orthologous genes.ConclusionDeep neural network architectures can efficiently predict whether a microbial gene is essential (or not) using only its sequence information.

Model-driven analysis of mutant fitness experiments improves genome-scale metabolic models of Zymomonas mobilis ZM4.

Genome-scale metabolic models have been utilized extensively in the study and engineering of the organisms they describe. Here we present the analysis of a published dataset from pooled transposon mutant fitness experiments as an approach for improving the accuracy and gene-reaction associations of a metabolic model for Zymomonas mobilis ZM4, an industrially relevant ethanologenic organism with extremely high glycolytic flux and low biomass yield. Gene essentiality predictions made by the draft model were compared to data from individual pooled mutant experiments to identify areas of the model requiring deeper validation. Subsequent experiments showed that some of the discrepancies between the model and dataset were caused by polar effects, mis-mapped barcodes, or mutants carrying both wild-type and transposon disrupted gene copies—highlighting potential limitations inherent to data from individual mutants in these high-throughput datasets. Therefore, we analyzed correlations in fitness scores across all 492 experiments in the dataset in the context of functionally related metabolic reaction modules identified within the model via flux coupling analysis. These correlations were used to identify candidate genes for a reaction in histidine biosynthesis lacking an annotated gene and highlight metabolic modules with poorly correlated gene fitness scores. Additional genes for reactions involved in biotin, ubiquinone, and pyridoxine biosynthesis in Z. mobilis were identified and confirmed using mutant complementation experiments. These discovered genes, were incorporated into the final model, iZM4_478, which contains 747 metabolic and transport reactions (of which 612 have gene-protein-reaction associations), 478 genes, and 616 unique metabolites, making it one of the most complete models of Z. mobilis ZM4 to date. The methods of analysis that we applied here with the Z. mobilis transposon mutant dataset, could easily be utilized to improve future genome-scale metabolic reconstructions for organisms where these, or similar, high-throughput datasets are available.

Expression-based prediction of human essential genes and candidate lncRNAs in cancer cells.

Essential genes are required for the reproductive success at either cellular or organismal level. The identification of essential genes is important for understanding the core biological processes and identifying effective therapeutic drug targets. However, experimental identification of essential genes is costly, time consuming and labor intensive. Although several machine learning models have been developed to predict essential genes, these models are not readily applicable to lncRNAs. Moreover, the currently available models cannot be used to predict essential genes in a specific cancer type. In this study, we have developed a new machine learning approach, XGEP (eXpression-based Gene Essentiality Prediction), to predict essential genes and candidate lncRNAs in cancer cells. The novelty of XGEP lies in the utilization of relevant features derived from the TCGA transcriptome dataset through collaborative embedding. When evaluated on the pan-cancer dataset, XGEP was able to accurately predict human essential genes and achieve significantly higher performance than previous models. Notably, several candidate lncRNAs selected by XGEP are reported to promote cell proliferation and inhibit cell apoptosis. Moreover, XGEP also demonstrated superior performance on cancer-type-specific datasets to identify essential genes. The comprehensive lists of candidate essential genes in specific cancer types may be used to guide experimental characterization and facilitate the discovery of drug targets for cancer therapy. The source code and datasets used in this study are freely available at https://github.com/BioDataLearning/XGEP. Supplementary data are available at Bioinformatics online.

Comparing the utility of in vivo transposon mutagenesis approaches in yeast species to infer gene essentiality

In vivo transposon mutagenesis, coupled with deep sequencing, enables large-scale genome-wide mutant screens for genes essential in different growth conditions. We analyzed six large-scale studies performed on haploid strains of three yeast species (Saccharomyces cerevisiae, Schizosaccaromyces pombe, and Candida albicans), each mutagenized with two of three different heterologous transposons (AcDs, Hermes, and PiggyBac). Using a machine-learning approach, we evaluated the ability of the data to predict gene essentiality. Important data features included sufficient numbers and distribution of independent insertion events. All transposons showed some bias in insertion site preference because of jackpot events, and preferences for specific insertion sequences and short-distance vs long-distance insertions. For PiggyBac, a stringent target sequence limited the ability to predict essentiality in genes with few or no target sequences. The machine learning approach also robustly predicted gene function in less well-studied species by leveraging cross-species orthologs. Finally, comparisons of isogenic diploid versus haploid S. cerevisiae isolates identified several genes that are haplo-insufficient, while most essential genes, as expected, were recessive. We provide recommendations for the choice of transposons and the inference of gene essentiality in genome-wide studies of eukaryotic haploid microbes such as yeasts, including species that have been less amenable to classical genetic studies.

Reconstruction and Validation of a Genome-Scale Metabolic Model of Streptococcus oralis (iCJ415), a Human Commensal and Opportunistic Pathogen.

The mitis group of streptococci (MGS) is a member of the healthy human microbiome in the oral cavity and upper respiratory tract. Troublingly, some MGS are able to escape this niche and cause infective endocarditis, a severe and devastating disease. Genome-scale models have been shown to be valuable in investigating metabolism of bacteria. Here we present the first genome-scale model, iCJ415, for Streptococcus oralis SK141. We validated the model using gene essentiality and amino acid auxotrophy data from closely related species. iCJ415 has 71-76% accuracy in predicting gene essentiality and 85% accuracy in predicting amino acid auxotrophy. Further, the phenotype of S. oralis was tested using the Biolog Phenotype microarrays, giving iCJ415 a 82% accuracy in predicting carbon sources. iCJ415 can be used to explore the metabolic differences within the MGS, and to explore the complicated metabolic interactions between different species in the human oral cavity.

Complete genome sequence and genome-scale metabolic modelling of Acinetobacter baumannii type strain ATCC 19606

Multidrug-resistant (MDR) Acinetobacter baumannii is a critical threat to global health. The type strain ATCC 19606 has been widely used in studying the virulence, pathogenesis and mechanisms of antimicrobial resistance in A. baumannii. However, the lack of a complete genome sequence is a hindrance towards detailed bioinformatic studies. Here we report the generation of a complete genome for ATCC 19606 using PacBio sequencing. ATCC 19606 genome consists of a 3,980,848-bp chromosome and a 9,450-bp plasmid pMAC, and harbours a chromosomal dihydropteroate synthase gene sul2 conferring resistance to sulphonamides and a plasmid-borne ohr gene conferring resistance to peroxides. The genome also contains 69 virulence genes involved in surface adherence, biofilm formation, extracellular phospholipase, iron uptake, immune evasion and quorum sensing. Insertion sequences ISCR2 and ISAba11 are embedded in a 36.1-Kb genomic island, suggesting an IS-mediated large-scale DNA recombination. Furthermore, a genome-scale metabolic model (GSMM) iATCC19606v2 was constructed using the complete genome annotation. The model iATCC19606v2 incorporated a periplasmic compartment, 1,422 metabolites, 2,114 reactions and 1,009 genes, and a set of protein crowding constraints taking into account enzyme abundance limitation. The prediction of bacterial growth on 190 carbon and 95 nitrogen sources achieved a high accuracy of 85.6% compared to Biolog experiment results. Based upon two transposon mutant libraries of AB5075 and ATCC 17978, the predictions of essential genes reached the accuracy of 87.6% and 82.1%, respectively. Together, the complete genome sequence and high-quality GSMM iATCC19606v2 provide valuable tools for antimicrobial systems pharmacological investigations on A. baumannii.

Predicting gene essentiality in Caenorhabditis elegans by feature engineering and machine-learning

Defining genes that are essential for life has major implications for understanding critical biological processes and mechanisms. Although essential genes have been identified and characterised experimentally using functional genomic tools, it is challenging to predict with confidence such genes from molecular and phenomic data sets using computational methods. Using extensive data sets available for the model organism Caenorhabditis elegans, we constructed here a machine-learning (ML)-based workflow for the prediction of essential genes on a genome-wide scale. We identified strong predictors for such genes and showed that trained ML models consistently achieve highly-accurate classifications. Complementary analyses revealed an association between essential genes and chromosomal location. Our findings reveal that essential genes in C. elegans tend to be located in or near the centre of autosomal chromosomes; are positively correlated with low single nucleotide polymorphim (SNP) densities and epigenetic markers in promoter regions; are involved in protein and nucleotide processing; are transcribed in most cells; are enriched in reproductive tissues or are targets for small RNAs bound to the argonaut CSR-1. Based on these results, we hypothesise an interplay between epigenetic markers and small RNA pathways in the germline, with transcription-based memory; this hypothesis warrants testing. From a technical perspective, further work is needed to evaluate whether the present ML-based approach will be applicable to other metazoans (including Drosophila melanogaster) for which comprehensive data sets (i.e. genomic, transcriptomic, proteomic, variomic, epigenetic and phenomic) are available.

Essential gene prediction in Drosophila melanogaster using machine learning approaches based on sequence and functional features

Genes are termed to be essential if their loss of function compromises viability or results in profound loss of fitness. On the genome scale, these genes can be determined experimentally employing RNAi or knockout screens, but this is very resource intensive. Computational methods for essential gene prediction can overcome this drawback, particularly when intrinsic (e.g. from the protein sequence) as well as extrinsic features (e.g. from transcription profiles) are considered. In this work, we employed machine learning to predict essential genes in Drosophila melanogaster. A total of 27,340 features were generated based on a large variety of different aspects comprising nucleotide and protein sequences, gene networks, protein-protein interactions, evolutionary conservation and functional annotations. Employing cross-validation, we obtained an excellent prediction performance. The best model achieved in D. melanogaster a ROC-AUC of 0.90, a PR-AUC of 0.30 and a F1 score of 0.34. Our approach considerably outperformed a benchmark method in which only features derived from the protein sequences were used (P < 0.001). Investigating which features contributed to this success, we found all categories of features, most prominently network topological, functional and sequence-based features. To evaluate our approach we performed the same workflow for essential gene prediction in human and achieved an ROC-AUC = 0.97, PR-AUC = 0.73, and F1 = 0.64.In summary, this study shows that using our well-elaborated assembly of features covering a broad range of intrinsic and extrinsic gene and protein features enabled intelligent systems to predict well the essentiality of genes in an organism.

Guiding the Refinement of Biochemical Knowledgebases with Ensembles of Metabolic Networks and Machine Learning.

SummaryMechanistic models explicitly represent hypothesized biological knowledge. As such, they offer more generalizability than data-driven models. However, identifying model curation efforts that improve performance for mechanistic models is nontrivial. Here, we develop a solution to this problem for genome-scale metabolic models. We generate an ensemble of models, each equally consistent with experimental data, then perform simulations with them. We apply machine learning to the simulation output to identify model structure variation that maximally influences simulations. These variants are high-priority candidates for curation through removal, addition, or reannotation in the model. We apply this approach, automated metabolic model ensemble-driven elimination of uncertainty with statistical learning (AMMEDEUS), to 29 bacterial species to improve gene essentiality predictions. We explore targets for individual species and compile pan-species targets to improve the database used during model construction. AMMEDEUS is an automated and performance-driven recommendation system that complements intuition during curation of biochemical knowledgebases.

Prediction of essential genes in prokaryote based on artificial neural network.

Rapid identification of new essential genes is necessary to understand biological mechanisms and identify potential targets for antimicrobial drugs. Many computational methods have been proposed. To construct an essential genes classifier which satisfies more different organisms, and to study the redundancy of features used in the prediction of essential genes. We designed a 57-12-1 artificial neural network model to predict the essential genes of 31 prokaryotic genomes. Four methods including self-predictions of each organism, the leave-one-genome-out method, predicting all by one organism, and self-predictions of all organisms were applied to assess the predictive performance. Additionally, the 57 features used in the artificial neural network model were analyzed by weighted principal component analysis to screen the key features strongly related to the essentiality of genes. Our results compared with previous researches indicate that our models had better generalizability. Furthermore, this method reduced the features to 29 while maintaining stable prediction performance overall, suggesting that some features are redundant for gene essentiality, and the screened features contained more important biological information for gene essentiality. This study showed the effectiveness and generalizability of our artificial neural network model. In addition, the screened features could be used as key features in computational analysis and biological experiments.

High-quality genome-scale metabolic modelling of Pseudomonas putida highlights its broad metabolic capabilities.

SummaryGenome‐scale reconstructions of metabolism are computational species‐specific knowledge bases able to compute systemic metabolic properties. We present a comprehensive and validated reconstruction of the biotechnologically relevant bacterium Pseudomonas putida KT2440 that greatly expands computable predictions of its metabolic states. The reconstruction represents a significant reactome expansion over available reconstructed bacterial metabolic networks. Specifically, iJN1462 (i) incorporates several hundred additional genes and associated reactions resulting in new predictive capabilities, including new nutrients supporting growth; (ii) was validated by in vivo growth screens that included previously untested carbon (48) and nitrogen (41) sources; (iii) yielded gene essentiality predictions showing large accuracy when compared with a knock‐out library and Bar‐seq data; and (iv) allowed mapping of its network to 82 P. putida sequenced strains revealing functional core that reflect the large metabolic versatility of this species, including aromatic compounds derived from lignin. Thus, this study provides a thoroughly updated metabolic reconstruction and new computable phenotypes for P. putida, which can be leveraged as a first step toward understanding the pan metabolic capabilities of Pseudomonas.

Metabolic Modeling of Streptococcus mutans Reveals Complex Nutrient Requirements of an Oral Pathogen.

Streptococcus mutans is a Gram-positive bacterium that thrives under acidic conditions and is a primary cause of tooth decay (dental caries). To better understand the metabolism of S. mutans on a systematic level, we manually constructed a genome-scale metabolic model of the S. mutans type strain UA159. The model, called iSMU, contains 675 reactions involving 429 metabolites and the products of 493 genes. We validated iSMU by comparing simulations with growth experiments in defined medium. The model simulations matched experimental results for 17 of 18 carbon source utilization assays and 47 of 49 nutrient depletion assays. We also simulated the effects of single gene deletions. The model's predictions agreed with 78.1% and 84.4% of the gene essentiality predictions from two experimental data sets. Our manually curated model is more accurate than S. mutans models generated from automated reconstruction pipelines and more complete than other manually curated models. We used iSMU to generate hypotheses about the S. mutans metabolic network. Subsequent genetic experiments confirmed that (i) S. mutans catabolizes sorbitol via a sorbitol-6-phosphate 2-dehydrogenase (SMU_308) and (ii) the Leloir pathway is required for growth on complex carbohydrates such as raffinose. We believe the iSMU model is an important resource for understanding the metabolism of S. mutans and guiding future experiments.IMPORTANCE Tooth decay is the most prevalent chronic disease in the United States. Decay is caused by the bacterium Streptococcus mutans, an oral pathogen that ferments sugars into tooth-destroying lactic acid. We constructed a complete metabolic model of S. mutans to systematically investigate how the bacterium grows. The model provides a valuable resource for understanding and targeting S. mutans' ability to outcompete other species in the oral microbiome.

Geptop 2.0: An Updated, More Precise, and Faster Geptop Server for Identification of Prokaryotic Essential Genes.

Geptop has performed effectively in the identification of prokaryotic essential genes since its first release in 2013. It estimates gene essentiality for prokaryotes based on orthology and phylogeny. Genome-scale essentiality data of more prokaryotic species are available, and the information has been collected into public essential gene repositories such as DEG and OGEE. A faster and more accurate toolkit is needed to meet the increasing prokaryotic genome data. We updated Geptop by supplementing more validated essentiality data into reference set (from 19 to 37 species), and introducing multi-process technology to accelerate the computing speed. Compared with Geptop 1.0 and other gene essentiality prediction models, Geptop 2.0 can generate more stable predictions and finish the computation in a shorter time. The software is available both as an online server and a downloadable standalone application. We hope that the improved Geptop 2.0 will facilitate researches in gene essentiality and the development of novel antibacterial drugs. The gene essentiality prediction tool is available at http://cefg.uestc.cn/geptop.

BOFdat: Generating biomass objective functions for genome-scale metabolic models from experimental data.

Genome-scale metabolic models (GEMs) are mathematically structured knowledge bases of metabolism that provide phenotypic predictions from genomic information. GEM-guided predictions of growth phenotypes rely on the accurate definition of a biomass objective function (BOF) that is designed to include key cellular biomass components such as the major macromolecules (DNA, RNA, proteins), lipids, coenzymes, inorganic ions and species-specific components. Despite its importance, no standardized computational platform is currently available to generate species-specific biomass objective functions in a data-driven, unbiased fashion. To fill this gap in the metabolic modeling software ecosystem, we implemented BOFdat, a Python package for the definition of a Biomass Objective Function from experimental data. BOFdat has a modular implementation that divides the BOF definition process into three independent modules defined here as steps: 1) the coefficients for major macromolecules are calculated, 2) coenzymes and inorganic ions are identified and their stoichiometric coefficients estimated, 3) the remaining species-specific metabolic biomass precursors are algorithmically extracted in an unbiased way from experimental data. We used BOFdat to reconstruct the BOF of the Escherichia coli model iML1515, a gold standard in the field. The BOF generated by BOFdat resulted in the most concordant biomass composition, growth rate, and gene essentiality prediction accuracy when compared to other methods. Installation instructions for BOFdat are available in the documentation and the source code is available on GitHub (https://github.com/jclachance/BOFdat).

Network-based methods for predicting essential genes or proteins: a survey.

Genes that are thought to be critical for the survival of organisms or cells are called essential genes. The prediction of essential genes and their products (essential proteins) is of great value in exploring the mechanism of complex diseases, the study of the minimal required genome for living cells and the development of new drug targets. As laboratory methods are often complicated, costly and time-consuming, a great many of computational methods have been proposed to identify essential genes/proteins from the perspective of the network level with the in-depth understanding of network biology and the rapid development of biotechnologies. Through analyzing the topological characteristics of essential genes/proteins in protein-protein interaction networks (PINs), integrating biological information and considering the dynamic features of PINs, network-based methods have been proved to be effective in the identification of essential genes/proteins. In this paper, we survey the advanced methods for network-based prediction of essential genes/proteins and present the challenges and directions for future research.

Metabolic Responses to Polymyxin Treatment in Acinetobacter baumannii ATCC 19606: Integrating Transcriptomics and Metabolomics with Genome-Scale Metabolic Modeling.

Multidrug-resistant (MDR) Acinetobacter baumannii has emerged as a very problematic pathogen over the past decades, with a high incidence in nosocomial infections. Discovered in the late 1940s but abandoned in the 1970s, polymyxins (i.e., polymyxin B and colistin) have been revived as the last-line therapy against Gram-negative "superbugs," including MDR A. baumannii. Worryingly, resistance to polymyxins in A. baumannii has been increasingly reported, urging the development of novel antimicrobial therapies to rescue this last-line class of antibiotics. In the present study, we integrated genome-scale metabolic modeling with multiomics data to elucidate the mechanisms of cellular responses to colistin treatment in A. baumannii. A genome-scale metabolic model, iATCC19606, was constructed for strain ATCC 19606 based on the literature and genome annotation, containing 897 genes, 1,270 reactions, and 1,180 metabolites. After extensive curation, prediction of growth on 190 carbon sources using iATCC19606 achieved an overall accuracy of 84.3% compared to Biolog experimental results. Prediction of gene essentiality reached a high accuracy of 86.1% and 82.7% compared to two transposon mutant libraries of AB5075 and ATCC 17978, respectively. Further integrative modeling with our correlative transcriptomics and metabolomics data deciphered the complex regulation on metabolic responses to colistin treatment, including (i) upregulated fluxes through gluconeogenesis, the pentose phosphate pathway, and amino acid and nucleotide biosynthesis; (ii) downregulated TCA cycle and peptidoglycan and lipopolysaccharide biogenesis; and (iii) altered fluxes over respiratory chain. Our results elucidated the interplay of multiple metabolic pathways under colistin treatment in A. baumannii and provide key mechanistic insights into optimizing polymyxin combination therapy. IMPORTANCE Combating antimicrobial resistance has been highlighted as a critical global health priority. Due to the drying drug discovery pipeline, polymyxins have been employed as the last-line therapy against Gram-negative "superbugs"; however, the detailed mechanisms of antibacterial killing remain largely unclear, hampering the improvement of polymyxin therapy. Our integrative modeling using the constructed genome-scale metabolic model iATCC19606 and the correlative multiomics data provide the fundamental understanding of the complex metabolic responses to polymyxin treatment in A. baumannii at the systems level. The model iATCC19606 may have a significant potential in antimicrobial systems pharmacology research in A. baumannii.