A Landscape of Pharmacogenomic Interactions in Cancer.

Systematic studies of cancer genomes have provided unprecedented insights into the molecular nature of cancer. Using this information to guide the development and application of therapies in the clinic is challenging. Here, we report how cancer-driven alterations identified in 11,289 tumors from 29 tissues (integrating somatic mutations, copy number alterations, DNA methylation, and gene expression) can be mapped onto 1,001 molecularly annotated human cancer cell lines and correlated with sensitivity to 265 drugs. We find that cell lines faithfully recapitulate oncogenic alterations identified in tumors, find that many of these associate with drug sensitivity/resistance, and highlight the importance of tissue lineage in mediating drug response. Logic-based modeling uncovers combinations of alterations that sensitize to drugs, while machine learning demonstrates the relative importance of different data types in predicting drug response. Our analysis and datasets are rich resources to link genotypes with cellular phenotypes and to identify therapeutic options for selected cancer sub-populations.

Metabolic reprogramming during Candida albicans planktonic-biofilm transition is modulated by the transcription factors Zcf15 and Zcf26.

Candida albicans is a commensal of the human microbiota that can form biofilms on implanted medical devices. These biofilms are tolerant to antifungals and to the host immune system. To identify novel genes modulating C. albicans biofilm formation, we performed a large-scale screen with 2,454 C. albicans doxycycline-dependent overexpression strains and identified 16 genes whose overexpression significantly hampered biofilm formation. Among those, overexpression of the ZCF15 and ZCF26 paralogs that encode transcription factors and have orthologs only in biofilm-forming species of the Candida clade, caused impaired biofilm formation both in vitro and in vivo. Interestingly, overexpression of ZCF15 impeded biofilm formation without any defect in hyphal growth. Transcript profiling, transcription factor binding, and phenotypic microarray analyses conducted upon overexpression of ZCF15 and ZCF26 demonstrated their role in reprogramming cellular metabolism by regulating central metabolism including glyoxylate and tricarboxylic acid cycle genes. Taken together, this study has identified a new set of biofilm regulators, including ZCF15 and ZCF26, that appear to control biofilm development through their specific role in metabolic remodeling.

Belt and braces: Two escape ways to maintain the cassette reservoir of large chromosomal integrons.

Integrons are adaptive devices that capture, stockpile, shuffle and express gene cassettes thereby sampling combinatorial phenotypic diversity. Some integrons called sedentary chromosomal integrons (SCIs) can be massive structures containing hundreds of cassettes. Since most of these cassettes are non-expressed, it is not clear how they remain stable over long evolutionary timescales. Recently, it was found that the experimental inversion of the SCI of Vibrio cholerae led to a dramatic increase of the cassette excision rate associated with a fitness defect. Here, we question the evolutionary sustainability of this apparently counter selected genetic context. Through experimental evolution, we find that the integrase is rapidly inactivated and that the inverted SCI can recover its original orientation by homologous recombination between two insertion sequences (ISs) present in the array. These two outcomes of SCI inversion restore the normal growth and prevent the loss of cassettes, enabling SCIs to retain their roles as reservoirs of functions. These results illustrate a nice interplay between gene orientation, genome rearrangement, bacterial fitness and demonstrate how integrons can benefit from their embedded ISs.

SARS-CoV-2-related bat virus behavior in human-relevant models sheds light on the origin of COVID-19.

Bat sarbecovirus BANAL-236 is highly related to SARS-CoV-2 and infects human cells, albeit lacking the furin cleavage site in its spike protein. BANAL-236 replicates efficiently and pauci-symptomatically in humanized mice and in macaques, where its tropism is enteric, strongly differing from that of SARS-CoV-2. BANAL-236 infection leads to protection against superinfection by a virulent strain. We find no evidence of antibodies recognizing bat sarbecoviruses in populations in close contact with bats in which the virus was identified, indicating that such spillover infections, if they occur, are rare. Six passages in humanized mice or in human intestinal cells, mimicking putative early spillover events, select adaptive mutations without appearance of a furin cleavage site and no change in virulence. Therefore, acquisition of a furin site in the spike protein is likely a pre-spillover event that did not occur upon replication of a SARS-CoV-2-like bat virus in humans or other animals. Other hypotheses regarding the origin of the SARS-CoV-2 should therefore be evaluated, including the presence of sarbecoviruses carrying a spike with a furin cleavage site in bats.

Reprohackathons: promoting reproducibility in bioinformatics through training.

MOTIVATION: The reproducibility crisis has highlighted the importance of improving the way bioinformatics data analyses are implemented, executed, and shared. To address this, various tools such as content versioning systems, workflow management systems, and software environment management systems have been developed. While these tools are becoming more widely used, there is still much work to be done to increase their adoption. The most effective way to ensure reproducibility becomes a standard part of most bioinformatics data analysis projects is to integrate it into the curriculum of bioinformatics Master's programs. RESULTS: In this article, we present the Reprohackathon, a Master's course that we have been running for the last 3 years at Université Paris-Saclay (France), and that has been attended by a total of 123 students. The course is divided into two parts. The first part includes lessons on the challenges related to reproducibility, content versioning systems, container management, and workflow systems. In the second part, students work on a data analysis project for 3-4 months, reanalyzing data from a previously published study. The Reprohackaton has taught us many valuable lessons, such as the fact that implementing reproducible analyses is a complex and challenging task that requires significant effort. However, providing in-depth teaching of the concepts and the tools during a Master's degree program greatly improves students' understanding and abilities in this area.

Experimental evolution links post-transcriptional regulation to Leishmania fitness gain.

The protozoan parasite Leishmania donovani causes fatal human visceral leishmaniasis in absence of treatment. Genome instability has been recognized as a driver in Leishmania fitness gain in response to environmental change or chemotherapy. How genome instability generates beneficial phenotypes despite potential deleterious gene dosage effects is unknown. Here we address this important open question applying experimental evolution and integrative systems approaches on parasites adapting to in vitro culture. Phenotypic analyses of parasites from early and late stages of culture adaptation revealed an important fitness tradeoff, with selection for accelerated growth in promastigote culture (fitness gain) impairing infectivity (fitness costs). Comparative genomics, transcriptomics and proteomics analyses revealed a complex regulatory network associated with parasite fitness gain, with genome instability causing highly reproducible, gene dosage-independent and -dependent changes. Reduction of flagellar transcripts and increase in coding and non-coding RNAs implicated in ribosomal biogenesis and protein translation were not correlated to dosage changes of the corresponding genes, revealing a gene dosage-independent, post-transcriptional mechanism of regulation. In contrast, abundance of gene products implicated in post-transcriptional regulation itself correlated to corresponding gene dosage changes. Thus, RNA abundance during parasite adaptation is controled by direct and indirect gene dosage changes. We correlated differential expression of small nucleolar RNAs (snoRNAs) with changes in rRNA modification, providing first evidence that Leishmania fitness gain in culture may be controlled by post-transcriptional and epitranscriptomic regulation. Our findings propose a novel model for Leishmania fitness gain in culture, where differential regulation of mRNA stability and the generation of modified ribosomes may potentially filter deleterious from beneficial gene dosage effects and provide proteomic robustness to genetically heterogenous, adapting parasite populations. This model challenges the current, genome-centric approach to Leishmania epidemiology and identifies the Leishmania transcriptome and non-coding small RNome as potential novel sources for the discovery of biomarkers that may be associated with parasite phenotypic adaptation in clinical settings.

Core genome sequencing and genotyping of Leptospira interrogans in clinical samples by target capture sequencing.

BACKGROUND: The life-threatening pathogen Leptospira interrogans is the most common agent of leptospirosis, an emerging zoonotic disease. However, little is known about the strains that are currently circulating worldwide due to the fastidious nature of the bacteria and the difficulty to isolate cultures. In addition, the paucity of bacteria in blood and other clinical samples has proven to be a considerable challenge for directly genotyping the agent of leptospirosis directly from patient material. Our understanding of the genetic diversity of strains during human infection is therefore limited. METHODS: Here, we carried out hybridization capture followed by Illumina sequencing of the core genome directly from 20 clinical samples that were PCR positive for pathogenic Leptospira to elucidate the genetic diversity of currently circulating Leptospira strains in mainland France. RESULTS: Capture with RNA probes covering the L. interrogans core genome resulted in a 72 to 13,000-fold increase in pathogen reads relative to standard sequencing without capture. Variant analysis of the genomes sequenced from the biological samples using 273 Leptospira reference genomes was then carried out to determine the genotype of the infecting strain. For samples with sufficient coverage (19/20 samples with coverage > 8×), we could unambiguously identify L. interrogans serovars Icterohaemorrhagiae and Copenhageni (14 samples), L. kirschneri serovar Grippotyphosa (4 samples), and L. interrogans serovar Pyrogenes (1 sample) as the infecting strains. CONCLUSIONS: We obtained high-quality genomic data with suitable coverage for confident core genome genotyping of the agent of leptospirosis for most of our clinical samples. The recovery of the genome of the serovars Icterohaemorrhagiae and Copenhageni directly from multiple clinical samples revealed low adaptive diversification of the core genes during human infection. The ability to generate culture-free genomic data opens new opportunities for better understanding of the epidemiology of this fastidious pathogen and pathogenesis of this neglected disease.

JC Polyomavirus Whole Genome Sequencing at the Single-Molecule Level Reveals Emerging Neurotropic Populations in Progressive Multifocal Leukoencephalopathy.

BACKGROUND: JC polyomavirus (JCV) mostly causes asymptomatic persistent renal infections but may give rise in immunosuppressed patients to neurotropic variants that replicate in the brain, causing progressive multifocal leukoencephalopathy (PML). Rearrangements in the JCV genome regulator noncoding control region (NCCR) and missense mutations in the viral capsid VP1 gene differentiate neurotropic variants from virus excreted in urine. METHODS: To investigate intrahost emergence of JCV neurotropic populations in PML, we deep sequenced JCV whole genome recovered from cerebrospinal fluid (CSF) and urine samples from 32 human immunodeficiency virus (HIV)-infected and non-HIV-infected PML patients at the single-molecule level. RESULTS: JCV strains distributed among 6 of 7 known genotypes. Common patterns of NCCR rearrangements included an initial deletion mostly located in a short 10-nucleotide sequence, followed by duplications/insertions. Multiple NCCR variants present in individual CSF samples shared at least 1 rearrangement, suggesting they stemmed from a unique viral population. NCCR variants independently acquired single or double PML-specific adaptive VP1 mutations. NCCR variants recovered from urine and CSF displayed opposite deletion or duplication patterns in binding sites for transcription factors. CONCLUSIONS: Long-read deep sequencing shed light on emergence of neurotropic JCV populations in PML.

Autotransporters Drive Biofilm Formation and Autoaggregation in the Diderm Firmicute Veillonella parvula.

The Negativicutes are a clade of the Firmicutes that have retained the ancestral diderm character and possess an outer membrane. One of the best studied Negativicutes, Veillonella parvula, is an anaerobic commensal and opportunistic pathogen inhabiting complex human microbial communities, including the gut and the dental plaque microbiota. Whereas the adhesion and biofilm capacities of V. parvula are expected to be crucial for its maintenance and development in these environments, studies of V. parvula adhesion have been hindered by the lack of efficient genetic tools to perform functional analyses in this bacterium. Here, we took advantage of a recently described naturally transformable V. parvula isolate, SKV38, and adapted tools developed for the closely related Clostridia spp. to perform random transposon and targeted mutagenesis to identify V. parvula genes involved in biofilm formation. We show that type V secreted autotransporters, typically found in diderm bacteria, are the main determinants of V. parvula autoaggregation and biofilm formation and compete with each other for binding either to cells or to surfaces, with strong consequences for V. parvula biofilm formation capacity. The identified trimeric autotransporters have an original structure compared to classical autotransporters identified in Proteobacteria, with an additional C-terminal domain. We also show that inactivation of the gene coding for a poorly characterized metal-dependent phosphohydrolase HD domain protein conserved in the Firmicutes and their closely related diderm phyla inhibits autotransporter-mediated biofilm formation. This study paves the way for further molecular characterization of V. parvula interactions with other bacteria and the host within complex microbiota environments.IMPORTANCEVeillonella parvula is an anaerobic commensal and opportunistic pathogen whose ability to adhere to surfaces or other bacteria and form biofilms is critical for it to inhabit complex human microbial communities such as the gut and oral microbiota. Although the adhesive capacity of V. parvula has been previously described, very little is known about the underlying molecular mechanisms due to a lack of genetically amenable Veillonella strains. In this study, we took advantage of a naturally transformable V. parvula isolate and newly adapted genetic tools to identify surface-exposed adhesins called autotransporters as the main molecular determinants of adhesion in this bacterium. This work therefore provides new insights on an important aspect of the V. parvula lifestyle, opening new possibilities for mechanistic studies of the contribution of biofilm formation to the biology of this major commensal of the oral-digestive tract.

Inferring causal molecular networks: empirical assessment through a community-based effort.

It remains unclear whether causal, rather than merely correlational, relationships in molecular networks can be inferred in complex biological settings. Here we describe the HPN-DREAM network inference challenge, which focused on learning causal influences in signaling networks. We used phosphoprotein data from cancer cell lines as well as in silico data from a nonlinear dynamical model. Using the phosphoprotein data, we scored more than 2,000 networks submitted by challenge participants. The networks spanned 32 biological contexts and were scored in terms of causal validity with respect to unseen interventional data. A number of approaches were effective, and incorporating known biology was generally advantageous. Additional sub-challenges considered time-course prediction and visualization. Our results suggest that learning causal relationships may be feasible in complex settings such as disease states. Furthermore, our scoring approach provides a practical way to empirically assess inferred molecular networks in a causal sense.

Sequanix: a dynamic graphical interface for Snakemake workflows.

Summary: We designed a PyQt graphical user interface-Sequanix-aimed at democratizing the use of Snakemake pipelines in the NGS space and beyond. By default, Sequanix includes Sequana NGS pipelines (Snakemake format) (http://sequana.readthedocs.io), and is also capable of loading any external Snakemake pipeline. New users can easily, visually, edit configuration files of expert-validated pipelines and can interactively execute these production-ready workflows. Sequanix will be useful to both Snakemake developers in exposing their pipelines and to a wide audience of users. Availability and implementation: Source on http://github.com/sequana/sequana, bio-containers on http://bioconda.github.io and Singularity hub (http://singularity-hub.org). Contact: dimitri.desvillechabrol@pasteur.fr or thomas.cokelaer@pasteur.fr. Supplementary information: Supplementary data are available at Bioinformatics online.

GDSCTools for mining pharmacogenomic interactions in cancer.

Motivation: Large pharmacogenomic screenings integrate heterogeneous cancer genomic datasets as well as anti-cancer drug responses on thousand human cancer cell lines. Mining this data to identify new therapies for cancer sub-populations would benefit from common data structures, modular computational biology tools and user-friendly interfaces. Results: We have developed GDSCTools: a software aimed at the identification of clinically relevant genomic markers of drug response. The Genomics of Drug Sensitivity in Cancer (GDSC) database (www.cancerRxgene.org) integrates heterogeneous cancer genomic datasets as well as anti-cancer drug responses on a thousand cancer cell lines. Including statistical tools (analysis of variance) and predictive methods (Elastic Net), as well as common data structures, GDSCTools allows users to reproduce published results from GDSC and to implement new analytical methods. In addition, non-GDSC data resources can also be analysed since drug responses and genomic features can be encoded as CSV files. Contact: thomas.cokelaer@pasteur.fr or saezrodriguez.rwth-aachen.de or mg12@sanger.ac.uk. Supplementary information: Supplementary data are available at Bioinformatics online.

Compatibility of 5-ethynyl-2'F-ANA UTP with in vitro selection for the generation of base-modified, nuclease resistant aptamers.

A modified nucleoside triphosphate bearing two modifications based on a 2'-deoxy-2'-fluoro-arabinofuranose sugar and a uracil nucleobase equipped with a C5-ethynyl moiety (5-ethynyl-2'F-ANA UTP) was synthesized. This nucleotide analog could enzymatically be incorporated into DNA oligonucleotides by primer extension and reverse transcribed to unmodified DNA. This nucleotide could be used in SELEX for the identification of high binding affinity and nuclease resistant aptamers.

Efficient randomization of biological networks while preserving functional characterization of individual nodes.

BACKGROUND: Networks are popular and powerful tools to describe and model biological processes. Many computational methods have been developed to infer biological networks from literature, high-throughput experiments, and combinations of both. Additionally, a wide range of tools has been developed to map experimental data onto reference biological networks, in order to extract meaningful modules. Many of these methods assess results' significance against null distributions of randomized networks. However, these standard unconstrained randomizations do not preserve the functional characterization of the nodes in the reference networks (i.e. their degrees and connection signs), hence including potential biases in the assessment. RESULTS: Building on our previous work about rewiring bipartite networks, we propose a method for rewiring any type of unweighted networks. In particular we formally demonstrate that the problem of rewiring a signed and directed network preserving its functional connectivity (F-rewiring) reduces to the problem of rewiring two induced bipartite networks. Additionally, we reformulate the lower bound to the iterations' number of the switching-algorithm to make it suitable for the F-rewiring of networks of any size. Finally, we present BiRewire3, an open-source Bioconductor package enabling the F-rewiring of any type of unweighted network. We illustrate its application to a case study about the identification of modules from gene expression data mapped on protein interaction networks, and a second one focused on building logic models from more complex signed-directed reference signaling networks and phosphoproteomic data. CONCLUSIONS: BiRewire3 it is freely available at https://www.bioconductor.org/packages/BiRewire/ , and it should have a broad application as it allows an efficient and analytically derived statistical assessment of results from any network biology tool.

Phosphoproteomic analyses reveal novel cross-modulation mechanisms between two signaling pathways in yeast.

Cells respond to environmental stimuli via specialized signaling pathways. Concurrent stimuli trigger multiple pathways that integrate information, predominantly via protein phosphorylation. Budding yeast responds to NaCl and pheromone via two mitogen-activated protein kinase cascades, the high osmolarity, and the mating pathways, respectively. To investigate signal integration between these pathways, we quantified the time-resolved phosphorylation site dynamics after pathway co-stimulation. Using shotgun mass spectrometry, we quantified 2,536 phosphopeptides across 36 conditions. Our data indicate that NaCl and pheromone affect phosphorylation events within both pathways, which thus affect each other at more levels than anticipated, allowing for information exchange and signal integration. We observed a pheromone-induced down-regulation of Hog1 phosphorylation due to Gpd1, Ste20, Ptp2, Pbs2, and Ptc1. Distinct Ste20 and Pbs2 phosphosites responded differently to the two stimuli, suggesting these proteins as key mediators of the information exchange. A set of logic models was then used to assess the role of measured phosphopeptides in the crosstalk. Our results show that the integration of the response to different stimuli requires complex interconnections between signaling pathways.

MEIGO: an open-source software suite based on metaheuristics for global optimization in systems biology and bioinformatics.

BACKGROUND: Optimization is the key to solving many problems in computational biology. Global optimization methods, which provide a robust methodology, and metaheuristics in particular have proven to be the most efficient methods for many applications. Despite their utility, there is a limited availability of metaheuristic tools. RESULTS: We present MEIGO, an R and Matlab optimization toolbox (also available in Python via a wrapper of the R version), that implements metaheuristics capable of solving diverse problems arising in systems biology and bioinformatics. The toolbox includes the enhanced scatter search method (eSS) for continuous nonlinear programming (cNLP) and mixed-integer programming (MINLP) problems, and variable neighborhood search (VNS) for Integer Programming (IP) problems. Additionally, the R version includes BayesFit for parameter estimation by Bayesian inference. The eSS and VNS methods can be run on a single-thread or in parallel using a cooperative strategy. The code is supplied under GPLv3 and is available at http://www.iim.csic.es/~gingproc/meigo.html. Documentation and examples are included. The R package has been submitted to BioConductor. We evaluate MEIGO against optimization benchmarks, and illustrate its applicability to a series of case studies in bioinformatics and systems biology where it outperforms other state-of-the-art methods. CONCLUSIONS: MEIGO provides a free, open-source platform for optimization that can be applied to multiple domains of systems biology and bioinformatics. It includes efficient state of the art metaheuristics, and its open and modular structure allows the addition of further methods.

BioServices: a common Python package to access biological Web Services programmatically.

MOTIVATION: Web interfaces provide access to numerous biological databases. Many can be accessed to in a programmatic way thanks to Web Services. Building applications that combine several of them would benefit from a single framework. RESULTS: BioServices is a comprehensive Python framework that provides programmatic access to major bioinformatics Web Services (e.g. KEGG, UniProt, BioModels, ChEMBLdb). Wrapping additional Web Services based either on Representational State Transfer or Simple Object Access Protocol/Web Services Description Language technologies is eased by the usage of object-oriented programming. AVAILABILITY AND IMPLEMENTATION: BioServices releases and documentation are available at http://pypi.python.org/pypi/bioservices under a GPL-v3 license.

Exhaustively characterizing feasible logic models of a signaling network using Answer Set Programming.

MOTIVATION: Logic modeling is a useful tool to study signal transduction across multiple pathways. Logic models can be generated by training a network containing the prior knowledge to phospho-proteomics data. The training can be performed using stochastic optimization procedures, but these are unable to guarantee a global optima or to report the complete family of feasible models. This, however, is essential to provide precise insight in the mechanisms underlaying signal transduction and generate reliable predictions. RESULTS: We propose the use of Answer Set Programming to explore exhaustively the space of feasible logic models. Toward this end, we have developed caspo, an open-source Python package that provides a powerful platform to learn and characterize logic models by leveraging the rich modeling language and solving technologies of Answer Set Programming. We illustrate the usefulness of caspo by revisiting a model of pro-growth and inflammatory pathways in liver cells. We show that, if experimental error is taken into account, there are thousands (11 700) of models compatible with the data. Despite the large number, we can extract structural features from the models, such as links that are always (or never) present or modules that appear in a mutual exclusive fashion. To further characterize this family of models, we investigate the input-output behavior of the models. We find 91 behaviors across the 11 700 models and we suggest new experiments to discriminate among them. Our results underscore the importance of characterizing in a global and exhaustive manner the family of feasible models, with important implications for experimental design. AVAILABILITY: caspo is freely available for download (license GPLv3) and as a web service at http://caspo.genouest.org/. SUPPLEMENTARY INFORMATION: Supplementary materials are available at Bioinformatics online. CONTACT: santiago.videla@irisa.fr.

Bile-induced biofilm formation in Bacteroides thetaiotaomicron requires magnesium efflux by an RND pump.

Bacteroides thetaiotaomicron is a prominent member of the human gut microbiota contributing to nutrient exchange, gut function, and maturation of the host's immune system. This obligate anaerobe symbiont can adopt a biofilm lifestyle, and it was recently shown that B. thetaiotaomicron biofilm formation is promoted by the presence of bile. This process also requires a B. thetaiotaomicron extracellular DNase, which is not, however, regulated by bile. Here, we showed that bile induces the expression of several Resistance-Nodulation-Division (RND) efflux pumps and that inhibiting their activity with a global competitive efflux inhibitor impaired bile-dependent biofilm formation. We then showed that, among the bile-induced RND-efflux pumps, only the tripartite BT3337-BT3338-BT3339 pump, re-named BipABC [for Bile Induced Pump A (BT3337), B (BT3338), and C (BT3339)], is required for biofilm formation. We demonstrated that BipABC is involved in the efflux of magnesium to the biofilm extracellular matrix, which leads to a decrease of extracellular DNA concentration. The release of magnesium in the biofilm matrix also impacts biofilm structure, potentially by modifying the electrostatic repulsion forces within the matrix, reducing interbacterial distance and allowing bacteria to interact more closely and form denser biofilms. Our study therefore, identified a new molecular determinant of B. thetaiotaomicron biofilm formation in response to bile salts and provides a better understanding on how an intestinal chemical cue regulates biofilm formation in a major gut symbiont.IMPORTANCEBacteroides thetaiotaomicron is a prominent member of the human gut microbiota able to degrade dietary and host polysaccharides, altogether contributing to nutrient exchange, gut function, and maturation of the host's immune system. This obligate anaerobe symbiont can adopt a biofilm community lifestyle, providing protection against environmental factors that might, in turn, protect the host from dysbiosis and dysbiosis-related diseases. It was recently shown that B. thetaiotaomicron exposure to intestinal bile promotes biofilm formation. Here, we reveal that a specific B. thetaiotaomicron membrane efflux pump is induced in response to bile, leading to the release of magnesium ions, potentially reducing electrostatic repulsion forces between components of the biofilm matrix. This leads to a reduction of interbacterial distance and strengthens the biofilm structure. Our study, therefore, provides a better understanding of how bile promotes biofilm formation in a major gut symbiont, potentially promoting microbiota resilience to stress and dysbiosis events.

Genomic and phenotypic characterization of Mycobacterium tuberculosis' closest-related non-tuberculous mycobacteria.

Four species of non-tuberculous mycobacteria (NTM) rated as biosafety level 1 or 2 (BSL-1/BSL-2) organisms and showing higher genomic similarity with Mycobacterium tuberculosis (Mtb) than previous comparator species Mycobacterium kansasii and Mycobacterium marinum were subjected to genomic and phenotypic characterization. These species named Mycobacterium decipiens, Mycobacterium lacus, Mycobacterium riyadhense, and Mycobacterium shinjukuense might represent "missing links" between low-virulent mycobacterial opportunists and the highly virulent obligate pathogen Mtb. We confirmed that M. decipiens is the closest NTM species to Mtb currently known and found that it has an optimal growth temperature of 32°C-35°C and not 37°C. M. decipiens showed resistance to rifampicin, isoniazid, and ethambutol, whereas M. lacus and M. riyadhense showed resistance to isoniazid and ethambutol. M. shinjukuense was sensitive to all three first-line TB drugs, and all four species were sensitive to bedaquiline, a third-generation anti-TB drug. Our results suggest these four NTM may be useful models for the identification and study of new anti-TB molecules, facilitated by their culture under non-BSL-3 conditions as compared to Mtb. M. riyadhense was the most virulent of the four species in cellular and mouse infection models. M. decipiens also multiplied in THP-1 cells at 35°C but was growth impaired at 37°C. Genomic comparisons showed that the espACD locus, essential for the secretion of ESX-1 proteins in Mtb, was present only in M. decipiens, which was able to secrete ESAT-6 and CFP-10, whereas secretion of these antigens varied in the other species, making the four species interesting examples for studying ESX-1 secretion mechanisms.IMPORTANCEIn this work, we investigated recently identified opportunistic mycobacterial pathogens that are genomically more closely related to Mycobacterium tuberculosis (Mtb) than previously used comparator species Mycobacterium kansasii and Mycobacterium marinum. We confirmed that Mycobacterium decipiens is the currently closest known species to the tubercle bacilli, represented by Mycobacterium canettii and Mtb strains. Surprisingly, the reference strain of Mycobacterium riyadhense (DSM 45176), which was purchased as a biosafety level 1 (BSL-1)-rated organism, was the most virulent of the four species in the tested cellular and mouse infection models, suggesting that a BSL-2 rating might be more appropriate for this strain than the current BSL-1 rating. Our work establishes the four NTM species as interesting study models to obtain new insights into the evolutionary mechanisms and phenotypic particularities of mycobacterial pathogens that likely have also impacted the evolution of the key pathogen Mtb.