Annotation-free prediction of microbial dioxygen utilization.

Aerobes require dioxygen (O2) to grow; anaerobes do not. However, nearly all microbes-aerobes, anaerobes, and facultative organisms alike-express enzymes whose substrates include O2, if only for detoxification. This presents a challenge when trying to assess which organisms are aerobic from genomic data alone. This challenge can be overcome by noting that O2 utilization has wide-ranging effects on microbes: aerobes typically have larger genomes encoding distinctive O2-utilizing enzymes, for example. These effects permit high-quality prediction of O2 utilization from annotated genome sequences, with several models displaying ≈80% accuracy on a ternary classification task for which blind guessing is only 33% accurate. Since genome annotation is compute-intensive and relies on many assumptions, we asked if annotation-free methods also perform well. We discovered that simple and efficient models based entirely on genomic sequence content-e.g., triplets of amino acids-perform as well as intensive annotation-based classifiers, enabling rapid processing of genomes. We further show that amino acid trimers are useful because they encode information about protein composition and phylogeny. To showcase the utility of rapid prediction, we estimated the prevalence of aerobes and anaerobes in diverse natural environments cataloged in the Earth Microbiome Project. Focusing on a well-studied O2 gradient in the Black Sea, we found quantitative correspondence between local chemistry (O2:sulfide concentration ratio) and the composition of microbial communities. We, therefore, suggest that statistical methods like ours might be used to estimate, or "sense," pivotal features of the chemical environment using DNA sequencing data.IMPORTANCEWe now have access to sequence data from a wide variety of natural environments. These data document a bewildering diversity of microbes, many known only from their genomes. Physiology-an organism's capacity to engage metabolically with its environment-may provide a more useful lens than taxonomy for understanding microbial communities. As an example of this broader principle, we developed algorithms that accurately predict microbial dioxygen utilization directly from genome sequences without annotating genes, e.g., by considering only the amino acids in protein sequences. Annotation-free algorithms enable rapid characterization of natural samples, highlighting quantitative correspondence between sequences and local O2 levels in a data set from the Black Sea. This example suggests that DNA sequencing might be repurposed as a multi-pronged chemical sensor, estimating concentrations of O2 and other key facets of complex natural settings.

Genome-wide screen identifies host loci that modulate Mycobacterium tuberculosis fitness in immunodivergent mice.

Genetic differences among mammalian hosts and among strains of Mycobacterium tuberculosis (Mtb) are well-established determinants of tuberculosis (TB) patient outcomes. The advent of recombinant inbred mouse panels and next-generation transposon mutagenesis and sequencing approaches has enabled dissection of complex host-pathogen interactions. To identify host and pathogen genetic determinants of Mtb pathogenesis, we infected members of the highly diverse BXD family of strains with a comprehensive library of Mtb transposon mutants (TnSeq). Members of the BXD family segregate for Mtb-resistant C57BL/6J (B6 or B) and Mtb-susceptible DBA/2J (D2 or D) haplotypes. The survival of each bacterial mutant was quantified within each BXD host, and we identified those bacterial genes that were differentially required for Mtb fitness across BXD genotypes. Mutants that varied in survival among the host family of strains were leveraged as reporters of "endophenotypes," each bacterial fitness profile directly probing specific components of the infection microenvironment. We conducted quantitative trait loci (QTL) mapping of these bacterial fitness endophenotypes and identified 140 host-pathogen QTL (hpQTL). We located a QTL hotspot on chromosome 6 (75.97-88.58 Mb) associated with the genetic requirement of multiple Mtb genes: Rv0127 (mak), Rv0359 (rip2), Rv0955 (perM), and Rv3849 (espR). Together, this screen reinforces the utility of bacterial mutant libraries as precise reporters of the host immunological microenvironment during infection and highlights specific host-pathogen genetic interactions for further investigation. To enable downstream follow-up for both bacterial and mammalian genetic research communities, all bacterial fitness profiles have been deposited into GeneNetwork.org and added into the comprehensive collection of TnSeq libraries in MtbTnDB.

Genome-wide screen identifies host loci that modulate M. tuberculosis fitness in immunodivergent mice.

Genetic differences among mammalian hosts and Mycobacterium tuberculosis ( Mtb ) strains determine diverse tuberculosis (TB) patient outcomes. The advent of recombinant inbred mouse panels and next-generation transposon mutagenesis and sequencing approaches has enabled dissection of complex host- pathogen interactions. To identify host and pathogen genetic determinants of Mtb pathogenesis, we infected members of the BXD family of mouse strains with a comprehensive library of Mtb transposon mutants (TnSeq). Members of the BXD family segregate for Mtb -resistant C57BL/6J (B6 or B ) and Mtb -susceptible DBA/2J (D2 or D ) haplotypes. The survival of each bacterial mutant was quantified within each BXD host, and we identified those bacterial genes that were differentially required for Mtb fitness across BXD genotypes. Mutants that varied in survival among the host family of strains were leveraged as reporters for "endophenotypes", each bacterial fitness profile directly probing specific components of the infection microenvironment. We conducted QTL mapping of these bacterial fitness endophenotypes and identified 140 h ost- p athogen quantitative trait loci ( hp QTL). We identified a QTL hotspot on chromosome 6 (75.97-88.58 Mb) associated with the genetic requirement of multiple Mtb genes; Rv0127 ( mak ), Rv0359 ( rip2 ), Rv0955 ( perM ), and Rv3849 ( espR ). Together, this screen reinforces the utility of bacterial mutant libraries as precise reporters of the host immunological microenvironment during infection and highlights specific host-pathogen genetic interactions for further investigation. To enable downstream follow-up for both bacterial and mammalian genetic research communities, all bacterial fitness profiles have been deposited into GeneNetwork.org and added into the comprehensive collection of TnSeq libraries in MtbTnDB.

Enzyme Substrate Prediction from Three-Dimensional Feature Representations Using Space-Filling Curves.

Compact and interpretable structural feature representations are required for accurately predicting properties and function of proteins. In this work, we construct and evaluate three-dimensional feature representations of protein structures based on space-filling curves (SFCs). We focus on the problem of enzyme substrate prediction, using two ubiquitous enzyme families as case studies: the short-chain dehydrogenase/reductases (SDRs) and the S-adenosylmethionine-dependent methyltransferases (SAM-MTases). Space-filling curves such as the Hilbert curve and the Morton curve generate a reversible mapping from discretized three-dimensional to one-dimensional representations and thus help to encode three-dimensional molecular structures in a system-independent way and with only a few adjustable parameters. Using three-dimensional structures of SDRs and SAM-MTases generated using AlphaFold2, we assess the performance of the SFC-based feature representations in predictions on a new benchmark database of enzyme classification tasks including their cofactor and substrate selectivity. Gradient-boosted tree classifiers yield binary prediction accuracy of 0.77-0.91 and area under curve (AUC) characteristics of 0.83-0.92 for the classification tasks. We investigate the effects of amino acid encoding, spatial orientation, and (the few) parameters of SFC-based encodings on the accuracy of the predictions. Our results suggest that geometry-based approaches such as SFCs are promising for generating protein structural representations and are complementary to the existing protein feature representations such as evolutionary scale modeling (ESM) sequence embeddings.

On scientific understanding with artificial intelligence.

An oracle that correctly predicts the outcome of every particle physics experiment, the products of every possible chemical reaction or the function of every protein would revolutionize science and technology. However, scientists would not be entirely satisfied because they would want to comprehend how the oracle made these predictions. This is scientific understanding, one of the main aims of science. With the increase in the available computational power and advances in artificial intelligence, a natural question arises: how can advanced computational systems, and specifically artificial intelligence, contribute to new scientific understanding or gain it autonomously? Trying to answer this question, we adopted a definition of 'scientific understanding' from the philosophy of science that enabled us to overview the scattered literature on the topic and, combined with dozens of anecdotes from scientists, map out three dimensions of computer-assisted scientific understanding. For each dimension, we review the existing state of the art and discuss future developments. We hope that this Perspective will inspire and focus research directions in this multidisciplinary emerging field.

Deciphering functional redundancy and energetics of malate oxidation in mycobacteria.

Oxidation of malate to oxaloacetate, catalyzed by either malate dehydrogenase (Mdh) or malate quinone oxidoreductase (Mqo), is a critical step of the tricarboxylic acid cycle. Both Mqo and Mdh are found in most bacterial genomes, but the level of functional redundancy between these enzymes remains unclear. A bioinformatic survey revealed that Mqo was not as widespread as Mdh in bacteria but that it was highly conserved in mycobacteria. We therefore used mycobacteria as a model genera to study the functional role(s) of Mqo and its redundancy with Mdh. We deleted mqo from the environmental saprophyte Mycobacterium smegmatis, which lacks Mdh, and found that Mqo was essential for growth on nonfermentable carbon sources. On fermentable carbon sources, the Δmqo mutant exhibited delayed growth and lowered oxygen consumption and secreted malate and fumarate as terminal end products. Furthermore, heterologous expression of Mdh from the pathogenic species Mycobacterium tuberculosis shortened the delayed growth on fermentable carbon sources and restored growth on nonfermentable carbon sources at a reduced growth rate. In M. tuberculosis, CRISPR interference of either mdh or mqo expression resulted in a slower growth rate compared to controls, which was further inhibited when both genes were knocked down simultaneously. These data reveal that exergonic Mqo activity powers mycobacterial growth under nonenergy limiting conditions and that endergonic Mdh activity complements Mqo activity, but at an energetic cost for mycobacterial growth. We propose Mdh is maintained in slow-growing mycobacterial pathogens for use under conditions such as hypoxia that require reductive tricarboxylic acid cycle activity.

Multiple acyl-CoA dehydrogenase deficiency kills Mycobacterium tuberculosis in vitro and during infection.

The human pathogen Mycobacterium tuberculosis depends on host fatty acids as a carbon source. However, fatty acid ß-oxidation is mediated by redundant enzymes, which hampers the development of antitubercular drugs targeting this pathway. Here, we show that rv0338c, which we refer to as etfD, encodes a membrane oxidoreductase essential for ß-oxidation in M. tuberculosis. An etfD deletion mutant is incapable of growing on fatty acids or cholesterol, with long-chain fatty acids being bactericidal, and fails to grow and survive in mice. Analysis of the mutant's metabolome reveals a block in ß-oxidation at the step catalyzed by acyl-CoA dehydrogenases (ACADs), which in other organisms are functionally dependent on an electron transfer flavoprotein (ETF) and its cognate oxidoreductase. We use immunoprecipitation to show that M. tuberculosis EtfD interacts with FixA (EtfB), a protein that is homologous to the human ETF subunit ß and is encoded in an operon with fixB, encoding a homologue of human ETF subunit α. We thus refer to FixA and FixB as EtfB and EtfA, respectively. Our results indicate that EtfBA and EtfD (which is not homologous to human EtfD) function as the ETF and oxidoreductase for ß-oxidation in M. tuberculosis and support this pathway as a potential target for tuberculosis drug development.

A thermodynamic atlas of carbon redox chemical space.

Redox biochemistry plays a key role in the transduction of chemical energy in living systems. However, the compounds observed in metabolic redox reactions are a minuscule fraction of chemical space. It is not clear whether compounds that ended up being selected as metabolites display specific properties that distinguish them from nonbiological compounds. Here, we introduce a systematic approach for comparing the chemical space of all possible redox states of linear-chain carbon molecules to the corresponding metabolites that appear in biology. Using cheminformatics and quantum chemistry, we analyze the physicochemical and thermodynamic properties of the biological and nonbiological compounds. We find that, among all compounds, aldose sugars have the highest possible number of redox connections to other molecules. Metabolites are enriched in carboxylic acid functional groups and depleted of ketones and aldehydes and have higher solubility than nonbiological compounds. Upon constructing the energy landscape for the full chemical space as a function of pH and electron-donor potential, we find that metabolites tend to have lower Gibbs energies than nonbiological molecules. Finally, we generate Pourbaix phase diagrams that serve as a thermodynamic atlas to indicate which compounds are energy minima in redox chemical space across a set of pH values and electron-donor potentials. While escape from thermodynamic equilibrium toward kinetically driven states is a hallmark of life and its origin, we envision that a deeper quantitative understanding of the environment-dependent thermodynamic landscape of putative prebiotic molecules will provide a crucial reference for future origins-of-life models.

Extreme Antagonism Arising from Gene-Environment Interactions.

Antagonistic interactions in biological systems, which occur when one perturbation blunts the effect of another, are typically interpreted as evidence that the two perturbations impact the same cellular pathway or function. Yet, this interpretation ignores extreme antagonistic interactions wherein an otherwise deleterious perturbation compensates for the function lost because of a prior perturbation. Here, we report on gene-environment interactions involving genetic mutations that are deleterious in a permissive environment but beneficial in a specific environment that restricts growth. These extreme antagonistic interactions constitute gene-environment analogs of synthetic rescues previously observed for gene-gene interactions. Our approach uses two independent adaptive evolution steps to address the lack of experimental methods to systematically identify such extreme interactions. We apply the approach to Escherichia coli by successively adapting it to defined glucose media without and with the antibiotic rifampicin. The approach identified multiple mutations that are beneficial in the presence of rifampicin and deleterious in its absence. The analysis of transcription shows that the antagonistic adaptive mutations repress a stringent response-like transcriptional program, whereas nonantagonistic mutations have an opposite transcriptional profile. Our approach represents a step toward the systematic characterization of extreme antagonistic gene-drug interactions, which can be used to identify targets to select against antibiotic resistance.

Two for the price of one: Attacking the energetic-metabolic hub of mycobacteria to produce new chemotherapeutic agents.

Cellular bioenergetics is an area showing promise for the development of new antimicrobials, antimalarials and cancer therapy. Enzymes involved in central carbon metabolism and energy generation are essential mediators of bacterial physiology, persistence and pathogenicity, lending themselves natural interest for drug discovery. In particular, succinate and malate are two major focal points in both the central carbon metabolism and the respiratory chain of Mycobacterium tuberculosis. Both serve as direct links between the citric acid cycle and the respiratory chain due to the quinone-linked reactions of succinate dehydrogenase, fumarate reductase and malate:quinone oxidoreductase. Inhibitors against these enzymes therefore hold the promise of disrupting two distinct, but essential, cellular processes at the same time. In this review, we discuss the roles and unique adaptations of these enzymes and critically evaluate the role that future inhibitors of these complexes could play in the bioenergetics target space.

A Mixed Quantum Chemistry/Machine Learning Approach for the Fast and Accurate Prediction of Biochemical Redox Potentials and Its Large-Scale Application to 315 000 Redox Reactions.

A quantitative understanding of the thermodynamics of biochemical reactions is essential for accurately modeling metabolism. The group contribution method (GCM) is one of the most widely used approaches to estimate standard Gibbs energies and redox potentials of reactions for which no experimental measurements exist. Previous work has shown that quantum chemical predictions of biochemical thermodynamics are a promising approach to overcome the limitations of GCM. However, the quantum chemistry approach is significantly more expensive. Here, we use a combination of quantum chemistry and machine learning to obtain a fast and accurate method for predicting the thermodynamics of biochemical redox reactions. We focus on predicting the redox potentials of carbonyl functional group reductions to alcohols and amines, two of the most ubiquitous carbon redox transformations in biology. Our method relies on semiempirical quantum chemistry calculations calibrated with Gaussian process (GP) regression against available experimental data and results in higher predictive power than the GCM at low computational cost. Direct calibration of GCM and fingerprint-based predictions (without quantum chemistry) with GP regression also results in significant improvements in prediction accuracy, demonstrating the versatility of the approach. We design and implement a network expansion algorithm that iteratively reduces and oxidizes a set of natural seed metabolites and demonstrate the high-throughput applicability of our method by predicting the standard potentials of more than 315 000 redox reactions involving approximately 70 000 compounds. Additionally, we developed a novel fingerprint-based framework for detecting molecular environment motifs that are enriched or depleted across different regions of the redox potential landscape. We provide open access to all source code and data generated.

Quantum chemistry reveals thermodynamic principles of redox biochemistry.

Thermodynamics dictates the structure and function of metabolism. Redox reactions drive cellular energy and material flow. Hence, accurately quantifying the thermodynamics of redox reactions should reveal design principles that shape cellular metabolism. However, only few redox potentials have been measured, and mostly with inconsistent experimental setups. Here, we develop a quantum chemistry approach to calculate redox potentials of biochemical reactions and demonstrate our method predicts experimentally measured potentials with unparalleled accuracy. We then calculate the potentials of all redox pairs that can be generated from biochemically relevant compounds and highlight fundamental trends in redox biochemistry. We further address the question of why NAD/NADP are used as primary electron carriers, demonstrating how their physiological potential range fits the reactions of central metabolism and minimizes the concentration of reactive carbonyls. The use of quantum chemistry can revolutionize our understanding of biochemical phenomena by enabling fast and accurate calculation of thermodynamic values.

MultiDK: A Multiple Descriptor Multiple Kernel Approach for Molecular Discovery and Its Application to Organic Flow Battery Electrolytes.

We propose a multiple descriptor multiple kernel (MultiDK) method for efficient molecular discovery using machine learning. We show that the MultiDK method improves both the speed and accuracy of molecular property prediction. We apply the method to the discovery of electrolyte molecules for aqueous redox flow batteries. Using multiple-type-as opposed to single-type-descriptors, we obtain more relevant features for machine learning. Following the principle of "wisdom of the crowds", the combination of multiple-type descriptors significantly boosts prediction performance. Moreover, by employing multiple kernels-more than one kernel function for a set of the input descriptors-MultiDK exploits nonlinear relations between molecular structure and properties better than a linear regression approach. The multiple kernels consist of a Tanimoto similarity kernel and a linear kernel for a set of binary descriptors and a set of nonbinary descriptors, respectively. Using MultiDK, we achieve an average performance of r2 = 0.92 with a test set of molecules for solubility prediction. We also extend MultiDK to predict pH-dependent solubility and apply it to a set of quinone molecules with different ionizable functional groups to assess their performance as flow battery electrolytes.

Quantum chemical approach to estimating the thermodynamics of metabolic reactions.

Thermodynamics plays an increasingly important role in modeling and engineering metabolism. We present the first nonempirical computational method for estimating standard Gibbs reaction energies of metabolic reactions based on quantum chemistry, which can help fill in the gaps in the existing thermodynamic data. When applied to a test set of reactions from core metabolism, the quantum chemical approach is comparable in accuracy to group contribution methods for isomerization and group transfer reactions and for reactions not including multiply charged anions. The errors in standard Gibbs reaction energy estimates are correlated with the charges of the participating molecules. The quantum chemical approach is amenable to systematic improvements and holds potential for providing thermodynamic data for all of metabolism.

High-resolution profiling of stationary-phase survival reveals yeast longevity factors and their genetic interactions.

Lifespan is influenced by a large number of conserved proteins and gene-regulatory pathways. Here, we introduce a strategy for systematically finding such longevity factors in Saccharomyces cerevisiae and scoring the genetic interactions (epistasis) among these factors. Specifically, we developed an automated competition-based assay for chronological lifespan, defined as stationary-phase survival of yeast populations, and used it to phenotype over 5,600 single- or double-gene knockouts at unprecedented quantitative resolution. We found that 14% of the viable yeast mutant strains were affected in their stationary-phase survival; the extent of true-positive chronological lifespan factors was estimated by accounting for the effects of culture aeration and adaptive regrowth. We show that lifespan extension by dietary restriction depends on the Swr1 histone-exchange complex and that a functional link between autophagy and the lipid-homeostasis factor Arv1 has an impact on cellular lifespan. Importantly, we describe the first genetic interaction network based on aging phenotypes, which successfully recapitulated the core-autophagy machinery and confirmed a role of the human tumor suppressor PTEN homologue in yeast lifespan and phosphatidylinositol phosphate metabolism. Our quantitative analysis of longevity factors and their genetic interactions provides insights into the gene-network interactions of aging cells.

Cell lineage analysis of the mammalian female germline.

Fundamental aspects of embryonic and post-natal development, including maintenance of the mammalian female germline, are largely unknown. Here we employ a retrospective, phylogenetic-based method for reconstructing cell lineage trees utilizing somatic mutations accumulated in microsatellites, to study female germline dynamics in mice. Reconstructed cell lineage trees can be used to estimate lineage relationships between different cell types, as well as cell depth (number of cell divisions since the zygote). We show that, in the reconstructed mouse cell lineage trees, oocytes form clusters that are separate from hematopoietic and mesenchymal stem cells, both in young and old mice, indicating that these populations belong to distinct lineages. Furthermore, while cumulus cells sampled from different ovarian follicles are distinctly clustered on the reconstructed trees, oocytes from the left and right ovaries are not, suggesting a mixing of their progenitor pools. We also observed an increase in oocyte depth with mouse age, which can be explained either by depth-guided selection of oocytes for ovulation or by post-natal renewal. Overall, our study sheds light on substantial novel aspects of female germline preservation and development.

Invariant distribution of promoter activities in Escherichia coli.

Cells need to allocate their limited resources to express a wide range of genes. To understand how Escherichia coli partitions its transcriptional resources between its different promoters, we employ a robotic assay using a comprehensive reporter strain library for E. coli to measure promoter activity on a genomic scale at high-temporal resolution and accuracy. This allows continuous tracking of promoter activity as cells change their growth rate from exponential to stationary phase in different media. We find a heavy-tailed distribution of promoter activities, with promoter activities spanning several orders of magnitude. While the shape of the distribution is almost completely independent of the growth conditions, the identity of the promoters expressed at different levels does depend on them. Translation machinery genes, however, keep the same relative expression levels in the distribution across conditions, and their fractional promoter activity tracks growth rate tightly. We present a simple optimization model for resource allocation which suggests that the observed invariant distributions might maximize growth rate. These invariant features of the distribution of promoter activities may suggest design constraints that shape the allocation of transcriptional resources.