Draft reconstruction.

The draft model was built with CarveMe version 1.5.1 using the annotated genome sequence of ATCC 17978 [39]. This was downloaded from the National Centre for Biotechnology Information (NCBI) at https://www.ncbi.nlm.nih.gov and has the assembly accession number ASM1542v1 [20]. Seven strain-specific assemblies are registered in NCBI; however, the chosen entry is also registered in Kyoto Encyclopedia of Genes and Genomes (KEGG) [38], which supports the model extension. The genome is 3.9 Mbp long and has two plasmids (pAB1 and pAB2). We set the SBML flavor to activate the extension for flux balance constraints (fbc) version 2 [98]. This extension enables semantic descriptions for domain-specific elements, such as metabolite chemical formulas and charges, along with reaction boundaries and gene-protein-reaction associations (GPRs). Moreover, the optional parameter gramneg provided CarveMe was selected to employ the specialized template for the Gram-negative bacteria. Compared to the Gram-positive template, the Gram-negative template comes with phosphatidylethanolamines, murein, and a lipopolysaccharide unit. Its biomass reaction involves membrane and cell wall components resulting in more accurate gene essentiality predictions in the lipid biosynthesis pathways.

Manual refinement and extension.

We started the manual refinement of the draft model by resolving syntactical errors within the Systems Biology Markup Language (SBML) [40] model file using the SBML Validator from the libSBML [46]. These errors involve improper file structure, incorrect or missing tags, missing or improperly formatted attributes, and invalid values. Missing metabolite charges and chemical formulas were retrieved from the Biochemical, Genetical, and Genomical (BiGG) [70] and ChEBI [99] databases, while mass- and charge-imbalanced reactions were corrected. The most intense part of the workflow is the manual network extension and gap-filling. This was done using the organism-specific databases KEGG [38] and BioCyc [100], together with ModelSEED [71]. We mapped the new gene locus tags to the old ones using the GenBank General Feature Format (GFF) [101] and added missing metabolic genes along with the respective reactions and metabolites into our model. The network’s connectivity was ensured by resolving as many dead-end (only produced but not consumed) and orphan (only consumed but not produced) metabolites as possible. Also, reactions with no connectivity were not included in the model, while reactions with no organism-specific gene evidence were removed from the model.

Erroneous energy generating cycles.

Energy-generating cycles (EGCs) are thermodynamically infeasible loops found in metabolic networks and have not been experimentally observed, unlike futile cycles. EGCs charge energy metabolites such as adenosine triphosphate (ATP) and uridine triphosphate (UTP) without any external source of nutrients, which may lead to incorrect and unrealistic energy increases. Their elimination is crucial for correcting the energy metabolism, as they can inflate the maximal biomass yields and render unreliable predictions. We checked their existence in iACB23LX applying an algorithm developed by Fritzemeier et al. [41].

We created a Python script that (1) defines and adds energy dissipation reactions (EDRs) in the network: where X is the metabolite of interest and (2) maximizes each EDR while blocking all influxes. This can be formulated as follows: subject to (1) where edr is the index of the current dissipation reaction, S is the stoichiometric matrix, the flux vector, E the set of all exchange reactions, and and the upper and lower bounds. The existence of EGCs is indicated by a positive optimal value of v_edr.

Totally we examined 14 energy metabolites: ATP, cytidine triphosphate (CTP), guanosine triphosphate (GTP), UTP, inosine triphosphate (ITP), nicotinamide adenine dinucleotide (NADH), nicotinamide adenine dinucleotide phosphate (NADPH), flavin adenine dinucleotide (FADH2), flavin mononucleotide (FMNH2), ubiquinol-8, menaquinol-8, demethylmenaquinol-8, acetyl-coenzyme A (CoA), and l-glutamate. Moreover, we tested the proton exchange between cytosol and periplasm.

In the case of existing EGCs, we examined the directionality and the gene evidence of all participated reactions using organism-specific information from BioCyc as reference [100].

Database annotations.

In this stage, the model was enriched with cross-linkings to various functional databases. Reactions and metabolites were annotated with relevant databases (e.g., KEGG [38], BRENDA [102], and UniProt [103]). These were included in the model as controlled vocabulary (CV) Terms following the Minimal Information Required In the Annotation of Models (MIRIAM) guidelines [104] and the resolution service at https://identifiers.org/. We used ModelPolisher [105] to complete the missing available metadata for all metabolites and genes. Similarly, metabolic genes were annotated with their KEGG [38], NCBI Protein, and RefSeq identifiers using the GFF file [101]. Systems Biology Ontology (SBO) terms were assigned to different model entities using a freely accessible standalone tool called SBOannotator [43]. The SBO terms are structured controlled vocabularies used in computational modeling to define and describe model entities unambiguously [106]. The SBOannotator was developed to automatically assign precise SBO terms to SBML models, mainly focusing on biochemical reactions, thereby enhancing the reproducibility and usability of biochemical networks [43]. In addition, Evidence and Conclusion Ontology (ECO) terms were added to every reaction to capture the type of evidence of biological assertions with BQB_IS_DESCRIBED BY as a biological qualifier. They are useful during quality control and mirror the curator’s confidence about the inclusion of a reaction. When multiple genes encode a single reaction, an ECO term was added for every participant gene. Both terms were incorporated into the final model.

Finally, reactions were annotated with the associated subsystems in which they participate using the KEGG [38] database and the biological qualifier BQB_OCCURS_IN. Moreover, the “groups” plugin was activated [48]. Every reaction that appeared in a given pathway was added as a groups:member, while each pathway was created as a group instance with sboTerm=“SBO:0000633” and groups:kind=“partonomy”.

Quality control and quality assurance.

The metabolic model testing (MEMOTE) [45] version 0.13.0 was used to assess and track the quality of our model after each modification, providing us with information regarding the model improvement. The Flux variability, Reaction deletion, Objective function, and Gene deletion (FROG) analysis framework was utilized to assess model reproducibility, ensuring reusability and results verification [47]. The resulting reports include comprehensive analyses of flux variability, reaction deletion, objective function, and gene deletion, providing a thorough evaluation of model performance. The final model was converted into the latest SBML Level 3 Version 2 [48] format using the libSBML package, while the SBML Validator tracked syntactical errors and ensured a valid format of the final model [46].

Strict aerobic growth check.

At the time of writing, CarveMe does not include reconstruction templates to differentiate between aerobic and anaerobic species. The directionality of reactions that produce or consume oxygen may affect the model’s ability to grow anaerobically. A. baumannii is defined to be a strictly aerobic species. Hence, we tested whether our model could grow with no oxygen supplementation. For this purpose, we examined all active oxygen-producing reactions under anaerobic conditions. We corrected their directionality based on the organism-specific information found in BioCyc [100] and kept only those with associated gene evidence.

Defining a minimal growth medium.

To determine the minimal number of nutrients needed for the bacterium to grow, we defined a minimal medium using iACB23LX. We determined the minimal amount of metabolites needed for growth using the M9 minimal medium (M9) (S1 Table) as a reference (iMinMed). We modeled growth on iMinMed by enabling the uptake of all metabolites that constitute the medium. The lower bound for the rest of the exchanges was set to 0 mmol/(g_DW · h). The final minimal medium is listed in Table 1 and in S1 Table. It consists of nine transition metals, a carbon source, a nitrogen source, a sulfur source, and a phosphorus source. The aerobic environment was simulated by setting the lower bound for the oxygen exchange to −10 mmol/(g_DW · h).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Composition of the computationally defined minimal growth medium, iMinMed.

It consists of nine transition metals, a carbon source, a nitrogen source, a sulfur source, and a phosphorus source. Oxygen was used to represent aerobic conditions.

https://doi.org/10.1371/journal.ppat.1012528.t001

Growth in chemically defined media.

We utilized experimentally verified growth media to examine the growth capabilities of iACB23LX. The LB medium serves as a common medium for the cultivation of A. baumannii. Consequently, we conducted an assessment of our model’s capacity to accurately simulate growth in this particular medium. Furthermore, we examined the growth of our model in the human nasal niche, considering that A. baumannii has been isolated from nasal samples within ICUs [10–12]. For this purpose, we utilized the SNM that imitates the human nasal habitat [97]. In all cases, if macromolecules or mixtures were present, we considered the constitutive molecular components for the medium definition. As our model was initially unable to reproduce growth on the applied media, we deployed the gap-filling option from CarveMe to detect missing reactions and gaps in the network [39]. All growth media formulations are available in S1 Table.

Rich medium definition.

To investigate our model’s growth rate when all nutrients are available to the bacterial cell, we defined the rich medium. For this purpose, we enabled the uptake of all extracellular metabolites by the model setting the lower bound of their exchange reactions to −10 mmol/(g_DW · h).

Evaluation of carbon and nitrogen utilization.

We employed the previously published Biolog Phenotypic Array data by Farrugia et al. for A. baumannii ATCC 17978 to validate the functionality of our model [26]. According to the experimental guidelines provided by Farrugia et al., we utilized M9 for all simulations [26]. The medium was then supplemented with d-xylose as a carbon source for the nitrogen testings, while ammonium served as the only nitrogen source for the carbon tests. As d-xylose was initially not part of the model, we conducted an extensive search in the organism-specific databases KEGG [38] and BioCyc [100] to include missing reactions.

The phenotypes were grouped by their maximal kinetic curve height. A trait was considered positive (“growth”) if the height exceeded the 115 and 101 OmniLog units for a nitrogen and carbon source, respectively. The prediction accuracy was evaluated by comparing the in silico-derived phenotypes to the Biolog results. More specifically, the overall model’s accuracy (ACC) was calculated by the overall agreement: (4) where true positive (TP) and true negative (TN) are correct predictions, while false positive (FP) and false negative (FN) are inconsistent predictions. Discrepancies were resolved via iterative manual model curation.

Gene perturbation analysis.

We performed in silico single-gene deletions on iACB23LX to detect essential genes. For this purpose, we utilized the single_gene_deletion function from the COBRApy [58] package. A gene is considered to be essential if a flux of 0.0 mmol/(g_DW · h) is predicted through the biomass reaction after setting the lower and upper bounds of the associated reaction(s) to 0.0 mmol/(g_DW · h).

Additionally, we examined the effect of gene deletions using two different optimization approaches: FBA [59] and minimization of metabolic adjustment (MOMA) [60]. Contrary to FBA, MOMA is based on quadratic programming, and the involved optimization problem is the Euclidean distance minimization in flux space. Moreover, it approximates the metabolic phenotype and relaxes the assumption of optimal growth flux for gene deletions [60].

The results were compared to gene essentiality data [57]. Wang et al. generated a random mutagenesis dataset including 15,000 unique transposon mutants using insertion sequencing (INSeq) [57]. Four additional transposon sequencing (Tn-seq) libraries of multiple A. baumannii isolates were employed to validate the model and increase confidence [57, 68, 77–79]. In this case only genes with an ortholog in ASM1542v1 were considered. Analogously to the experimental settings, the nutrient uptake constraints were set to define the LB medium. From the 453 genes experimentally identified as essential, 191 could be compared to our predictions. The rest were not part of iACB23LX due to their non-metabolic functions. To measure the effect of a single deletion, we calculated the fold change (FC) between the model’s growth rate after (gr_KO) and before (gr_WT) a single knockout. This is formulated as follows: (5)

To this end, if FC_gr = 0, the deleted gene is classified as essential, meaning its removal prevented the network from producing at least one key biomass metabolite predicting no growth. Similarly, if FC_gr = 1, the deletion of the gene from the network did not affect the growth phenotype (labeled as inessential), while when 0 < FC_gr < 1, the removal of this gene affected partially the biomass production (labeled as partially essential). The complete lists of the gene essentiality results are available in S2 and S3 Tables.

To explore the potential of in silico identified essential genes as new drug candidates against A. baumannii infections, we probed the queries of predicted false negative candidates against the human proteome using Basic Local Alignment Search Tool (BLAST) [107]. The protein sequences were aligned to the human protein sequences using the default settings of the NCBI BLASTp tool (word size: 6, matrix: BLOSUM62, gap costs: 11 for existence and 1 for extension). To eliminate adverse effects and ensure no interference with human-like proteins, queries with any non-zero alignment score with the human proteome were not considered. Lastly, we searched the DrugBank database version 5.1.9 to find inhibitors or ligands known to act with the enzymes encoded by the non-homologous genes [66].

Prediction and experimental validation of bacterial growth on various nutritional environments.

Constraint-based modeling approaches, such as FBA, estimate flux rates indicating how metabolites flow through the metabolic network and predict cellular phenotypes for various growth scenarios. A. baumannii is known to be strictly aerobic and, compared to the majority of Acinetobacter species, it is not considered ubiquitous in nature. As a nosocomial pathogen, it has been mostly detected in hospital environments, particularly in the ICUs, and within the human nasal microbiota [10–12]. We examined various growth conditions to ensure that iACB23LX recapitulates these already known and fundamental phenotypes.

First, we tested our model’s capability to simulate a strictly aerobic growth. For this purpose, we examined the directionality of all active oxygen-producing and -consuming reactions when the oxygen uptake was disabled (see S1 Fig). We observed an accumulation of periplasmic oxygen by reactions that carried remarkably high fluxes, leading to growth even when oxygen import was turned off. We examined each reaction individually and removed those without gene evidence to correct this. More specifically, we removed the periplasmic catalase (CATpp), one of bacteria’s main hydrogen peroxide scavengers. This enzyme is typically active in the cytosol [49] and was not part of any precursor A. baumannii GEM or was found only in cytosol (iLP844 [31]). To fill the gap and enable the usage of the periplasmic hydrogen peroxide, we added the (PEAMNOpp) in the model. Eventually, iACB23LX demonstrated growth only in the presence of oxygen using a rich medium (all exchange reactions are open).

Furthermore, we determined the minimal number of metabolites necessary for growth using iACB23LX and the M9 as a reference. Minimal growth media typically consist of carbon, nitrogen, phosphorus, and sulfur sources, as well multiple inorganic salts and transition metals. These metals are crucial for the growth and survival of all three domains of life; however, they can be transformed into toxic compounds in hyper-availability [50]. The exact composition of our minimal medium (iMinMed) is shown in Table 1. It comprises nine transition metals, acetate as the carbon source, ammonium as a nitrogen source, sulfate as a sulfur source, and phosphate as a phosphorus source. Previous studies have highlighted the importance of nutrient metals for A. baumannii to survive within the host. More specifically, the bacterium utilizes these metals as co-factors for vital cellular processes [51]. Manganese and zinc have also been studied as essential determinants of host defense against A. baumannii-acquired pneumonia through their sequestering by calprotectin via a type of bonding called chelation [52]. In the computational simulations, growth rates below 2.81 mmol/(g_DW · h) were considered realistic. This threshold corresponds to the doubling time of the fastest-growing organism, Vibrio natriegens, which is 14.8 minutes [45]. Table 2 displays the predicted growth rates of iACB23LX in the respective culture media. In LB, our model simulated with the highest growth rate; 0.6065 mmol/(g_DW · h). With our self-defined minimal medium, iMinMed, our model exhibited the lowest rate; 0.5503 mmol/(g_DW · h). Notably, our experimental validation revealed similarity between the growth rates obtained from the in vitro respiratory curves (Table 2 and S2 Fig) and those predicted by our in silico simulations. Additionally, we examined the growth rate of our model in a rich medium, in which all nutrients were available to the model. With this, the flux through the biomass production was the highest, 2.1858 mmol/(g_DW · h), as expected. This is still less than the growth rate of the fastest organism, increasing the confidence in our model’s consistency and simulation capabilities. Initially, iACB23LX could not predict any realistic growth rate for the simulated media. Using the gap-filling function of CarveMe [39], we detected three enzymes (PHPYROX, OXADC, and LCYSTAT) whose addition into the metabolic network resulted in successful growth in all tested media.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Simulated and empirical growth rates of ATCC 17978 in various growth media.

The tested media are the computationally-defined minimal medium (iMinMed), the LB, and the SNM. Computational growth rates are given in mmol/(g_DW · h), while in vitro rates in h⁻¹. Doubling times are calculated in minutes. The media formulations are available in S1 Table.

https://doi.org/10.1371/journal.ppat.1012528.t002

Functional validation of iACB23LX using nutrient utilization data.

Multiple in silico approaches have hitherto been employed to predict lethal genes and to assess growth metrics on different carbon/nitrogen sources for severe pathogenic organisms including Mycobacterium tuberculosis [53, 54] and Staphylococcus aureus [55, 56]. In 2013 and 2014, two studies were published that examined the catabolic phenome and gene essentialities of the strain ATCC 17978 [26, 57]. We used these datasets to evaluate the overall performance (functionality and accuracy) of iACB23LX.

Our first validation experiment assessed the accuracy of our model’s carbon and nitrogen catabolism potentials using the large-scale phenotypic data provided by Farrugia et al. [26]. While the authors tested a larger number of compounds overall, we were only able to examine 80 carbon sources and 48 nitrogen sources. For the remaining molecules, either no BiGG identifier existed, or they were not part of the metabolic network. Following the experimental protocol by Farrugia et al., we applied the M9 medium and enabled d-xylose as the sole carbon source for the nitrogen testings. As d-xylose was initially not part of the reconstructed network, we conducted extensive literature and database search to include associated missing reactions. This improved the prediction accuracy, especially for the carbon sources, where an amelioration of 19% was achieved. In more detail, despite the comprehensive manual curation, the first draft model was reconstructed using the automated tool CarveMe [39]. This resulted in the incorrect inclusion of transport reactions, which were consequently removed to reduce the number of false positive predictions. In both cases, our main objective was to improve the accuracy while keeping the number of orphan and dead-end metabolites low and removing only reactions with no gene evidence (lack of assigned GPR). Similarly, missing reactions were identified and included in the network to eliminate the false negative predictions. For instance, in accordance with the phenotypic data, ATCC 17978 should not be able to grow when utilizing d-trehalose as the sole carbon source. Our model initially predicted a growth phenotype for this carbon source. To overcome this conflict, we deleted the reaction TREP with no organism-specific gene evidence, i.e., no assigned GPR. However, it was not feasible to resolve all inconsistencies since adding transport reactions to resolve false positives or false negatives in the nitrogen testings led to more false predictions in the carbon sources. More specifically, when adenosine, inosine, l-homoserine, and uridine are utilized as sole carbon sources, the model should not predict growth, while as sole nitrogen sources they should result in a non-zero objective value. In this case, adding transporters would resolve false predictions in the nitrogen tests, while it would have induced more false predictions in the carbon sources tests. Altogether, iACB23LX exhibited an overall accuracy of 86.3% for the carbon and 79.2% for the nitrogen sources test (Fig 4c). By adding their corresponding transport reactions, we resolved discrepancies regarding uridine, inosine, adenosine, and l-homoserine. Our model was able to catabolize 49 sole carbon and 40 sole nitrogen sources (see Fig 4a and 4b), recapitulating totally 69 and 38 experimentally-derived phenotypes, respectively.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 4. Model predictions compared to the Biolog experimental measurements for various carbon and nitrogen sources.

From the Biolog data, only substances mappable to model metabolites were included, while the M9 medium was applied. (a) and (b) The model’s ability to catabolize various carbon and nitrogen sources was assessed using the strain-specific phenotypic data by Farrugia et al. [26]. Grey indicates no growth, and orange indicates growth. Totally, 80 and 48 compounds were tested as sole carbon and nitrogen sources, respectively. Out of these, 69 and 38 phenotypes were recapitulated successfully by iACB23LX. (c) Confusion matrices of model predictions and Biolog experimental measurements. The overall accuracy of iACB23LX is 86.3% for the carbon (left matrix) and 79.2% for the nitrogen (right matrix) testings. Orange represents correct predictions, and grey represents wrong predictions.

https://doi.org/10.1371/journal.ppat.1012528.g004

We further assessed the ability of iACB23LX to predict known gene essentialities. First, 1,164 in silico single gene deletions were conducted on both LB and rich growth media, respectively to identify all lethal gene deletions. Subsequently, the ratio between the growth rate after and before the respective knockouts (FC_gr) was calculated, and the genes were classified accordingly (see Materials and methods). For the optimization, two mathematics-based approaches from the COBRApy [58] package were deployed: the FBA [59] and the MOMA [60]. Between the two methods, a similar distribution of the FC_gr values was observed (Fig 5a and 5b). Using FBA, 97, 75, and 991 genes were predicted to be essential, partially essential, and inessential on LB, respectively. Similarly, optimization with MOMA resulted in 110, 85, and 968 genes (Fig 5c and S3 and S2 and S3 Tables). These genes were primarily associated with the biosynthesis of cofactors and vitamins, the amino acid/nucleotide metabolism, the energy metabolism, and the metabolism of terpenoids and polyketides. Additionally, we examined in more detail how nutrition availability impacts the gene essentiality by conducting single-gene knockouts in the rich medium. Both optimization methods yielded a higher number of essential genes when the model had to adapt its metabolic behavior due to lacking nutrients, i.e., with LB, compared to the rich medium (Fig 5c and S2 and S3 Tables). In general, FBA detected more genes to be dispensable for growth in both nutritional environments. On the other hand, MOMA classified more genes as essential or partially essential (Fig 5c and S2 and S3 Tables), while genes from FBA build a subset of the essential genes derived by MOMA. Furthermore, we validated the prediction accuracy of iACB23LX using gene essentiality data. First, we analyzed the transposon mutant library developed by Wang et al. as it examines the same A. baumannii genome that we used to build our model [57].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Fig 5. Gene essentiality analysis using iACB23LX.

(a) and (b) Distribution of the FC_gr values calculated for all genes included in iACB23LX. Red lines represent FBA predictions and grey are ratios derived with MOMA. Totally 1,164 knockouts were conducted using each method in LB and rich media. (c) Classification of gene essentialities in essential, inessential, and partially essential based on their FC_gr values. (d) Accuracy of gene essentiality predictions based on empirical data. The in silico results were compared to the Wang et al. transposon library [57]. The LB medium was applied to mirror the experimental settings. The metabolic network exhibited 87% accuracy with FBA (left) and MOMA (right). Beige indicates correct predictions; grey indicates incorrect predictions. (e) Comparative analysis of essential genes predicted by iACB23LX versus those identified in multiple Tn-seq studies.

https://doi.org/10.1371/journal.ppat.1012528.g005

Using this dataset and the LB medium, our model demonstrated a prediciton accuracy of 87% with both optimization methods (Fig 5d). To enhance the robustness and generalizability of our model across diverse datasets, we compared our model’s predictions to four additional Tn-seq datasets. A total of 43 genes were labeled as essential in all studied Tn-seq data and were predicted as essential by our model. The derived predictive accuracies ranged between 86.6% and 88.9% (Fig 5e), underscoring the efficacy of our model across diverse high-throughput gene essentiality datasets. We further analyzed the predicted false negative genes and probed their proteomes to investigate the existence of human homologs (see Materials and methods and S4 Table). Our aim was to remove cross-linkings to human-similar proteins, as pathways or enzymes absent in humans are valuable sources of druggable targets against infectious diseases [61]. Out of the 37 genes that our model predicted to be essential using FBA and MOMA, which contradicted the experimental results, 17 were found to be non-homologous to human (S4 Table). Genes with a zero similarity score were defined as non-homologous, while those with a non-zero similarity score were excluded to avoid targeting human-like proteins. Some examples are the genes encoding the enolpyruvylshikimate phosphate (EPSP) synthase (A1S_2276), chorismate synthase (A1S_1694), riboflavin synthase (A1S_0223), phosphogluconate dehydratase (A1S_0483), dihydrofolate reductase (DHFR) (A1S_0457), and 2-keto-3-deoxy-6-phosphogluconate (KDPG) aldolase (A1S 0484). The EPSP synthase converts the shikimate-3-phosphate together with phosphoenolpyruvate to 5-o-(1-carboxyvinyl)-3-phosphoshikimic acid. Subsequently, the chorismate synthase catalyses the conversion of the 5-o-(1-carboxyvinyl)-3-phosphoshikimic acid to chorismate, the seventh and last step within the shikimate pathway [62]. Chorismate is the common precursor in the production of the aromatic compounds tryptophan, phenylalanine, and tyrosine, as well as folate and menaquinones during the bacterial life cycle. The shikimate pathway is of particular interest due to its absence from the human host metabolome and its vital role in bacterial metabolism and virulence. Moreover, the enzyme riboflavin synthase catalyses the final step of riboflavin (vitamin B2) biosynthesis with no participating cofactors. Riboflavin can be produced by most microorganisms compared to humans, who have to externally uptake them via food supplements. Also, it plays an important role in the growth of different microbes, especially due to its photosynthesizing property that marks it as a non-invasive and safe therapeutic strategy against bacterial infections [63]. Lastly, the phosphogluconate dehydratase catalyses the dehydration of 6-phospho-d-gluconate to KDPG, the precursor of pyruvate and 3-phospho-d-glycerate [64]. This enzyme is part of the Entner–Doudoroff pathway that catabolizes glucose to pyruvate, similarly to glycolysis, but using a different set of enzymes [65].

We further assessed the druggability of our essential non-homologous proteins and investigated the existence of inhibitors or compounds known to interact with the enzymes. For this, we used the online DrugBank database that contains detailed information on various drugs and drug targets [66]. In all cases, the listed drugs are of unknown pharmacological action, and there is still no evidence indicating the enzymes’ association with the molecule’s mechanism of action. For instance, the flavin mononucleotide and the cobalt hexamine ion were listed as known inhibitors of yet unknown function against the chorismate synthase, while glyphosate, shikimate-3-phosphate, and formic acid have been experimentally found to act with EPSP synthase. Six non-homologous genes were marked as hypothetical or putative in the KEGG [38] database and/or lacked enzyme-associated information. We searched for drug leads by aligning the query sequences against the DrugBank’s database to find homologous proteins. Two out of six were found to have a protein hit. More specifically, the protein encoded by A1S_0589 was found to have high sequence identity with the phosphocarrier protein HPr of Enterococcus faecalis (Bit-score: 48.5), while the translation product of A1S_0706 resembles the sugar phosphatase YbiV of Escherichia coli (Bit-score: 225.3). According to DrugBank, dexfosfoserine and aspartate beryllium trifluoride have been experimentally determined to bind to these enzymes; however, their pharmacological action is still unknown. The S4 Table lists all non-homologous essential genes reported for iACB23LX.

Overall, iACB23LX exhibits high agreement to all validation tests and can, therefore, be used to systematically derive associations between genotypes and phenotypes.