An antibiotic target ranking and prioritization pipeline combining sequence, structure and network-based approaches exemplified for Serratia marcescens

GuiltyTargets: Prioritization of Novel Therapeutic Targets With Network Representation Learning

The majority of clinical trials fail due to low efficacy of investigated drugs, often resulting from a poor choice of target protein. Existing computational approaches aim to support target selection either via genetic evidence or by putting potential targets into the context of a disease specific network reconstruction. The purpose of this work was to investigate whether network representation learning techniques could be used to allow for a machine learning based prioritization of putative targets. We propose a novel target prioritization approach, GuiltyTargets, which relies on attributed network representation learning of a genome-wide protein-protein interaction network annotated with disease-specific differential gene expression and uses positive-unlabeled (PU) machine learning for candidate ranking. We evaluated our approach on 12 datasets from six diseases of different type (cancer, metabolic, neurodegenerative) within a 10 times repeated 5-fold stratified cross-validation and achieved AUROC values between 0.92 - 0.97, significantly outperforming previous approaches that relied on manually engineered topological features. Moreover, we showed that GuiltyTargets allows for target repositioning across related disease areas. An application of GuiltyTargets to Alzheimer's disease resulted in a number of highly ranked candidates that are currently discussed as targets in the literature. Interestingly, one (COMT) is also the target of an approved drug (Tolcapone) for Parkinson's disease, highlighting the potential for target repositioning with our method. The GuiltyTargets Python package is available on PyPI and all code used for analysis can be found under the MIT License at https://github.com/GuiltyTargets. Attributed network representation learning techniques provide an interesting approach to effectively leverage the existing knowledge about the molecular mechanisms in different diseases. In this work, the combination with positive-unlabeled learning for target prioritization demonstrated a clear superiority compared to classical feature engineering approaches. Our work highlights the potential of attributed network representation learning for target prioritization. Given the overarching relevance of networks in computational biology we believe that attributed network representation learning techniques could have a broader impact in the future.

Geptop 2.0: Accurately Select Essential Genes from the List of Protein-Coding Genes in Prokaryotic Genomes

Computational tool composites alternative way to identify essential genes and it is low-cost and time-efficient. Based on experimental essentiality sets deposited in the databases DEG and OGEE as reference, we developed an automatically computational tool named Geptop to select essential genes from the set of protein-coding genes in a prokaryotic genome, which utilizes the strategy of reciprocally best hit for homology search and evolutionary distance for weight assigning. The latest version of Geptop is 2.0 ( http://guolab.whu.edu.cn/geptop ), which can predict gene essentiality with the mean AUC 0f 0.84 in prokaryotes and is more stable. The chapter is to briefly introduce the tool and tell how to use it.

Alveolar Regeneration in COVID-19 Patients: A Network Perspective

A viral infection involves entry and replication of viral nucleic acid in a host organism, subsequently leading to biochemical and structural alterations in the host cell. In the case of SARS-CoV-2 viral infection, over-activation of the host immune system may lead to lung damage. Albeit the regeneration and fibrotic repair processes being the two protective host responses, prolonged injury may lead to excessive fibrosis, a pathological state that can result in lung collapse. In this review, we discuss regeneration and fibrosis processes in response to SARS-CoV-2 and provide our viewpoint on the triggering of alveolar regeneration in coronavirus disease 2019 (COVID-19) patients.

An Ebola, Neisseria and Trypanosoma human protein interaction census reveals a conserved human protein cluster targeted by various human pathogens

Filovirus ebolavirus (ZE; Zaire ebolavirus, Bundibugyo ebolavirus), Neisseria meningitidis (NM), and Trypanosoma brucei (Tb) are serious infectious pathogens, spanning viruses, bacteria and protists and all may target the blood and central nervous system during their life cycle. NM and Tb are extracellular pathogens while ZE is obligatory intracellular, targetting immune privileged sites. By using interactomics and comparative evolutionary analysis we studied whether conserved human proteins are targeted by these pathogens. We examined 2797 unique pathogen-targeted human proteins. The information derived from orthology searches of experimentally validated protein-protein interactions (PPIs) resulted both in unique and shared PPIs for each pathogen. Comparing and analyzing conserved and pathogen-specific infection pathways for NM, TB and ZE, we identified human proteins predicted to be targeted in at least two of the compared host-pathogen networks. However, four proteins were common to all three host-pathogen interactomes: the elongation factor 1-alpha 1 (EEF1A1), the SWI/SNF complex subunit SMARCC2 (matrix-associated actin-dependent regulator of chromatin subfamily C), the dolichyl-diphosphooligosaccharide--protein glycosyltransferase subunit 1 (RPN1), and the tubulin beta-5 chain (TUBB). These four human proteins all are also involved in cytoskeleton and its regulation and are often addressed by various human pathogens. Specifically, we found (i) 56 human pathogenic bacteria and viruses that target these four proteins, (ii) the well researched new pandemic pathogen SARS-CoV-2 targets two of these four human proteins and (iii) nine human pathogenic fungi (yet another evolutionary distant organism group) target three of the conserved proteins by 130 high confidence interactions.

Genome-wide inference of the Camponotus floridanus protein-protein interaction network using homologous mapping and interacting domain profile pairs

Apart from some model organisms, the interactome of most organisms is largely unidentified. High-throughput experimental techniques to determine protein-protein interactions (PPIs) are resource intensive and highly susceptible to noise. Computational methods of PPI determination can accelerate biological discovery by identifying the most promising interacting pairs of proteins and by assessing the reliability of identified PPIs. Here we present a first in-depth study describing a global view of the ant Camponotus floridanus interactome. Although several ant genomes have been sequenced in the last eight years, studies exploring and investigating PPIs in ants are lacking. Our study attempts to fill this gap and the presented interactome will also serve as a template for determining PPIs in other ants in future. Our C. floridanus interactome covers 51,866 non-redundant PPIs among 6,274 proteins, including 20,544 interactions supported by domain-domain interactions (DDIs), 13,640 interactions supported by DDIs and subcellular localization, and 10,834 high confidence interactions mediated by 3,289 proteins. These interactions involve and cover 30.6% of the entire C. floridanus proteome.

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.

Identification of Antifungal Targets Based on Computer Modeling

Aspergillus fumigatus is a saprophytic, cosmopolitan fungus that attacks patients with a weak immune system. A rational solution against fungal infection aims to manipulate fungal metabolism or to block enzymes essential for Aspergillus survival. Here we discuss and compare different bioinformatics approaches to analyze possible targeting strategies on fungal-unique pathways. For instance, phylogenetic analysis reveals fungal targets, while domain analysis allows us to spot minor differences in protein composition between the host and fungi. Moreover, protein networks between host and fungi can be systematically compared by looking at orthologs and exploiting information from host⁻pathogen interaction databases. Further data—such as knowledge of a three-dimensional structure, gene expression data, or information from calculated metabolic fluxes—refine the search and rapidly put a focus on the best targets for antimycotics. We analyzed several of the best targets for application to structure-based drug design. Finally, we discuss general advantages and limitations in identification of unique fungal pathways and protein targets when applying bioinformatics tools.

Comparative subtractive proteomics based ranking for antibiotic targets against the dirtiest superbug: Acinetobacter baumannii

Multidrug-resistant Acinetobacter baumannii is indeed to be the most successful nosocomial pathogen responsible for myriad infections in modern health care system. Computational methodologies based on genomics and proteomics proved to be powerful tools for providing substantial information about different aspects of A. baumannii biology that made it possible to design new approaches for treating multi, extensive and total drug resistant isolates of A. baumannii. In this current approach, 35 completely annotated proteomes of A. bauamnnii were filtered through a comprehensive subtractive proteomics pipeline for broad-spectrum drug candidates. In total, 10 proteins (KdsA, KdsB, LpxA, LpxC, LpxD, GpsE, PhoB, UvrY, KdpE and OmpR) could serve as ideal candidates for designing novel antibiotics. The work was extended with KdsA enzyme for structure information, prediction of intrinsic disorders, active site details, and structure based virtual screening of library containing natural product-like scaffolds. Most of the enzyme structure has fixed three-dimensional conformation. The selection of inhibitor for KdsA enzyme was based on druglikeness, pharmacokinetics and docking scores. Compound-4636 (5-((3-chloro-5-methyl-2-phenyl-2,3-dihydro-1H-pyrazol-4-yl)methoxy)-2-(((1-hydroxy-4-methylpentan-2-yl)amino)methyl)phenol) was revealed as the most potent inhibitor against A. baumannii KdsA enzyme having Gold fitness score of 77.68 and Autodock binding energy of -6.2 kcal/mol. The inhibitor completely follows Lipinski rule of five, Ghose rule, and Egan rule. Molecular dynamics simulation for KdsA and KdsA-4636 complex was performed for 100 ns to unveil what conformational changes the enzyme underwent in the absence and presence of the inhibitor, respectively. The average root means square deviation (RMSD) for both systems was found 3.5 Å, which signifies stable structure of the enzyme in both bounded and unbounded states. Absolute binding energy using Molecular Mechanics-Generalized Born Surface Area (MM-GBSA) reflected high affinity and vigorous interactions of the inhibitor with enzyme active residues. Findings of the current study could open up new avenues for experimentalists to design new potent antibiotics by targeting the targets screened in this study.