Discovery of antimicrobial peptides in the global microbiome with machine learning.

Novel antibiotics are urgently needed to combat the antibiotic-resistance crisis. We present a machine-learning-based approach to predict antimicrobial peptides (AMPs) within the global microbiome and leverage a vast dataset of 63,410 metagenomes and 87,920 prokaryotic genomes from environmental and host-associated habitats to create the AMPSphere, a comprehensive catalog comprising 863,498 non-redundant peptides, few of which match existing databases. AMPSphere provides insights into the evolutionary origins of peptides, including by duplication or gene truncation of longer sequences, and we observed that AMP production varies by habitat. To validate our predictions, we synthesized and tested 100 AMPs against clinically relevant drug-resistant pathogens and human gut commensals both in vitro and in vivo. A total of 79 peptides were active, with 63 targeting pathogens. These active AMPs exhibited antibacterial activity by disrupting bacterial membranes. In conclusion, our approach identified nearly one million prokaryotic AMP sequences, an open-access resource for antibiotic discovery.

Disentangling Genetic and Environmental Effects on the Proteotypes of Individuals.

Proteotypes, like genotypes, have been found to vary between individuals in several studies, but consistent molecular functional traits across studies remain to be quantified. In a meta-analysis of 11 proteomics datasets from humans and mice, we use co-variation of proteins in known functional modules across datasets and individuals to obtain a consensus landscape of proteotype variation. We find that individuals differ considerably in both protein complex abundances and stoichiometry. We disentangle genetic and environmental factors impacting these metrics, with genetic sex and specific diets together explaining 13.5% and 11.6% of the observed variation of complex abundance and stoichiometry, respectively. Sex-specific differences, for example, include various proteins and complexes, where the respective genes are not located on sex-specific chromosomes. Diet-specific differences, added to the individual genetic backgrounds, might become a starting point for personalized proteotype modulation toward desired features.

Unravelling the collateral damage of antibiotics on gut bacteria.

Antibiotics are used to fight pathogens but also target commensal bacteria, disturbing the composition of gut microbiota and causing dysbiosis and disease1. Despite this well-known collateral damage, the activity spectrum of different antibiotic classes on gut bacteria remains poorly characterized. Here we characterize further 144 antibiotics from a previous screen of more than 1,000 drugs on 38 representative human gut microbiome species2. Antibiotic classes exhibited distinct inhibition spectra, including generation dependence for quinolones and phylogeny independence for ß-lactams. Macrolides and tetracyclines, both prototypic bacteriostatic protein synthesis inhibitors, inhibited nearly all commensals tested but also killed several species. Killed bacteria were more readily eliminated from in vitro communities than those inhibited. This species-specific killing activity challenges the long-standing distinction between bactericidal and bacteriostatic antibiotic classes and provides a possible explanation for the strong effect of macrolides on animal3-5 and human6,7 gut microbiomes. To mitigate this collateral damage of macrolides and tetracyclines, we screened for drugs that specifically antagonized the antibiotic activity against abundant Bacteroides species but not against relevant pathogens. Such antidotes selectively protected Bacteroides species from erythromycin treatment in human-stool-derived communities and gnotobiotic mice. These findings illluminate the activity spectra of antibiotics in commensal bacteria and suggest strategies to circumvent their adverse effects on the gut microbiota.

Combinatorial, additive and dose-dependent drug-microbiome associations.

During the transition from a healthy state to cardiometabolic disease, patients become heavily medicated, which leads to an increasingly aberrant gut microbiome and serum metabolome, and complicates biomarker discovery1-5. Here, through integrated multi-omics analyses of 2,173 European residents from the MetaCardis cohort, we show that the explanatory power of drugs for the variability in both host and gut microbiome features exceeds that of disease. We quantify inferred effects of single medications, their combinations as well as additive effects, and show that the latter shift the metabolome and microbiome towards a healthier state, exemplified in synergistic reduction in serum atherogenic lipoproteins by statins combined with aspirin, or enrichment of intestinal Roseburia by diuretic agents combined with beta-blockers. Several antibiotics exhibit a quantitative relationship between the number of courses prescribed and progression towards a microbiome state that is associated with the severity of cardiometabolic disease. We also report a relationship between cardiometabolic drug dosage, improvement in clinical markers and microbiome composition, supporting direct drug effects. Taken together, our computational framework and resulting resources enable the disentanglement of the effects of drugs and disease on host and microbiome features in multimedicated individuals. Furthermore, the robust signatures identified using our framework provide new hypotheses for drug-host-microbiome interactions in cardiometabolic disease.

SPIRE: a Searchable, Planetary-scale mIcrobiome REsource.

Meta'omic data on microbial diversity and function accrue exponentially in public repositories, but derived information is often siloed according to data type, study or sampled microbial environment. Here we present SPIRE, a Searchable Planetary-scale mIcrobiome REsource that integrates various consistently processed metagenome-derived microbial data modalities across habitats, geography and phylogeny. SPIRE encompasses 99 146 metagenomic samples from 739 studies covering a wide array of microbial environments and augmented with manually-curated contextual data. Across a total metagenomic assembly of 16 Tbp, SPIRE comprises 35 billion predicted protein sequences and 1.16 million newly constructed metagenome-assembled genomes (MAGs) of medium or high quality. Beyond mapping to the high-quality genome reference provided by proGenomes3 (http://progenomes.embl.de), these novel MAGs form 92 134 novel species-level clusters, the majority of which are unclassified at species level using current tools. SPIRE enables taxonomic profiling of these species clusters via an updated, custom mOTUs database (https://motu-tool.org/) and includes several layers of functional annotation, as well as crosslinks to several (micro-)biological databases. The resource is accessible, searchable and browsable via http://spire.embl.de.

proGenomes3: approaching one million accurately and consistently annotated high-quality prokaryotic genomes.

The interpretation of genomic, transcriptomic and other microbial 'omics data is highly dependent on the availability of well-annotated genomes. As the number of publicly available microbial genomes continues to increase exponentially, the need for quality control and consistent annotation is becoming critical. We present proGenomes3, a database of 907 388 high-quality genomes containing 4 billion genes that passed stringent criteria and have been consistently annotated using multiple functional and taxonomic databases including mobile genetic elements and biosynthetic gene clusters. proGenomes3 encompasses 41 171 species-level clusters, defined based on universal single copy marker genes, for which pan-genomes and contextual habitat annotations are provided. The database is available at http://progenomes.embl.de/.

Opportunities and barriers in omics-based biomarker discovery for steatotic liver diseases.

The rising prevalence of liver diseases related to obesity and excessive use of alcohol is fuelling an increasing demand for accurate biomarkers aimed at community screening, diagnosis of steatohepatitis and significant fibrosis, monitoring, prognostication and prediction of treatment efficacy. Breakthroughs in omics methodologies and the power of bioinformatics have created an excellent opportunity to apply technological advances to clinical needs, for instance in the development of precision biomarkers for personalised medicine. Via omics technologies, biological processes from the genes to circulating protein, as well as the microbiome - including bacteria, viruses and fungi, can be investigated on an axis. However, there are important barriers to omics-based biomarker discovery and validation, including the use of semi-quantitative measurements from untargeted platforms, which may exhibit high analytical, inter- and intra-individual variance. Standardising methods and the need to validate them across diverse populations presents a challenge, partly due to disease complexity and the dynamic nature of biomarker expression at different disease stages. Lack of validity causes lost opportunities when studies fail to provide the knowledge needed for regulatory approvals, all of which contributes to a delayed translation of these discoveries into clinical practice. While no omics-based biomarkers have matured to clinical implementation, the extent of data generated has enabled the hypothesis-free discovery of a plethora of candidate biomarkers that warrant further validation. To explore the many opportunities of omics technologies, hepatologists need detailed knowledge of commonalities and differences between the various omics layers, and both the barriers to and advantages of these approaches.

The Menin story in acute myeloid leukaemia-The road to success.

The treatment of acute myeloid leukaemia (AML) has changed fundamentally in the last decade with many new targeted therapies entering clinics. Some of the most interesting agents under development are Menin inhibitors which interfere with the interaction of Menin with wild-type (wt) KMT2A or a KMT2A-fusion protein and thereby downregulate the leukaemic gene expression (MEIS1, PBX3, HOX) in NPM1 mutant or KMT2A-rearranged leukaemia. Other HOX and MEIS1 expressing leukaemias may also be sensitive to Menin inhibition. Following the encouraging results as monotherapy in refractory and relapsed AML, the combination of Menin inhibitors with chemotherapeutic agents and other targeted drugs is being investigated clinically.

Sphingolipids Are Depleted in Alcohol-Related Liver Fibrosis.

BACKGROUND & AIMS: Alcohol disturbs hepatic lipid synthesis and transport, but the role of lipid dysfunction in alcohol-related liver disease (ALD) is unclear. In this biopsy-controlled, prospective, observational study, we characterized the liver and plasma lipidomes in patients with early ALD. METHODS: We performed mass spectrometry-based lipidomics of paired liver and plasma samples from 315 patients with ALD and of plasma from 51 matched healthy controls. We associated lipid levels with histologic fibrosis, inflammation, and steatosis with correction for multiple testing and adjustment for confounders. We further investigated sphingolipid regulation by means of quantitative real-time polymerase chain reaction sequencing of microRNA, prediction of liver-related events, and tested causality with Mendelian randomization. RESULTS: We detected 198 lipids in the liver and 236 lipids in the circulation from 18 lipid classes. Most sphingolipids (sphingomyelins and ceramides) and phosphocholines were co-down-regulated in both liver and plasma, where lower abundance correlated with higher fibrosis stage. Sphingomyelins showed the most pronounced negative correlation to fibrosis, mirrored by negative correlations in both liver and plasma with hepatic inflammation. Reduced sphingomyelins predicted future liver-related events. This seemed to be characteristic of "pure ALD," as sphingomyelin levels were higher in patients with concomitant metabolic syndrome and ALD/nonalcoholic fatty liver disease overlap. Mendelian randomization in FinnGen and UK Biobanks indicated ALD as the cause of low sphingomyelins, and alcohol use disorder did not correlate with genetic susceptibility to low sphingomyelin levels. CONCLUSIONS: Alcohol-related liver fibrosis is characterized by selective and progressive lipid depletion in liver and blood, particularly sphingomyelins, which also associates with progression to liver-related events.

Consistency across multi-omics layers in a drug-perturbed gut microbial community.

Multi-omics analyses are used in microbiome studies to understand molecular changes in microbial communities exposed to different conditions. However, it is not always clear how much each omics data type contributes to our understanding and whether they are concordant with each other. Here, we map the molecular response of a synthetic community of 32 human gut bacteria to three non-antibiotic drugs by using five omics layers (16S rRNA gene profiling, metagenomics, metatranscriptomics, metaproteomics and metabolomics). We find that all the omics methods with species resolution are highly consistent in estimating relative species abundances. Furthermore, different omics methods complement each other for capturing functional changes. For example, while nearly all the omics data types captured that the antipsychotic drug chlorpromazine selectively inhibits Bacteroidota representatives in the community, the metatranscriptome and metaproteome suggested that the drug induces stress responses related to protein quality control. Metabolomics revealed a decrease in oligosaccharide uptake, likely caused by Bacteroidota depletion. Our study highlights how multi-omics datasets can be utilized to reveal complex molecular responses to external perturbations in microbial communities.

Extensive impact of non-antibiotic drugs on human gut bacteria.

A few commonly used non-antibiotic drugs have recently been associated with changes in gut microbiome composition, but the extent of this phenomenon is unknown. Here, we screened more than 1,000 marketed drugs against 40 representative gut bacterial strains, and found that 24% of the drugs with human targets, including members of all therapeutic classes, inhibited the growth of at least one strain in vitro. Particular classes, such as the chemically diverse antipsychotics, were overrepresented in this group. The effects of human-targeted drugs on gut bacteria are reflected on their antibiotic-like side effects in humans and are concordant with existing human cohort studies. Susceptibility to antibiotics and human-targeted drugs correlates across bacterial species, suggesting common resistance mechanisms, which we verified for some drugs. The potential risk of non-antibiotics promoting antibiotic resistance warrants further exploration. Our results provide a resource for future research on drug-microbiome interactions, opening new paths for side effect control and drug repurposing, and broadening our view of antibiotic resistance.

Immunoproteasome function maintains oncogenic gene expression in KMT2A-complex driven leukemia.

Pharmacologic targeting of chromatin-associated protein complexes has shown significant responses in KMT2A-rearranged (KMT2A-r) acute myeloid leukemia (AML) but resistance frequently develops to single agents. This points to a need for therapeutic combinations that target multiple mechanisms. To enhance our understanding of functional dependencies in KMT2A-r AML, we have used a proteomic approach to identify the catalytic immunoproteasome subunit PSMB8 as a specific vulnerability. Genetic and pharmacologic inactivation of PSMB8 results in impaired proliferation of murine and human leukemic cells while normal hematopoietic cells remain unaffected. Disruption of immunoproteasome function drives an increase in transcription factor BASP1 which in turn represses KMT2A-fusion protein target genes. Pharmacologic targeting of PSMB8 improves efficacy of Menin-inhibitors, synergistically reduces leukemia in human xenografts and shows preserved activity against Menin-inhibitor resistance mutations. This identifies and validates a cell-intrinsic mechanism whereby selective disruption of proteostasis results in altered transcription factor abundance and repression of oncogene-specific transcriptional networks. These data demonstrate that the immunoproteasome is a relevant therapeutic target in AML and that targeting the immunoproteasome in combination with Menin-inhibition could be a novel approach for treatment of KMT2A-r AML.

Development of a quantitative proteomics approach for cyclooxygenases and lipoxygenases in parallel to quantitative oxylipin analysis allowing the comprehensive investigation of the arachidonic acid cascade.

Oxylipins derived from the cyclooxygenase (COX) and lipoxygenase (LOX) pathways of the arachidonic acid (ARA) cascade are essential for the regulation of the inflammatory response and many other physiological functions. Comprehensive analytical methods comprised of oxylipin and protein abundance analysis are required to fully understand mechanisms leading to changes within these pathways. Here, we describe the development of a quantitative multi-omics approach combining liquid chromatography tandem mass spectrometry-based targeted oxylipin metabolomics and proteomics. As the first targeted proteomics method to cover these pathways, it enables the quantitative analysis of all human COX (COX-1 and COX-2) and relevant LOX pathway enzymes (5-LOX, 12-LOX, 15-LOX, 15-LOX-2, and FLAP) in parallel to the analysis of 239 oxylipins with our targeted oxylipin metabolomics method from a single sample. The detailed comparison between MRM3 and classical MRM-based detection in proteomics showed increased selectivity for MRM3, while MRM performed better in terms of sensitivity (LLOQ, 16-122 pM vs. 75-840 pM for the same peptides), linear range (up to 1.5-7.4 µM vs. 4-368 nM), and multiplexing capacities. Thus, the MRM mode was more favorable for this pathway analysis. With this sensitive multi-omics approach, we comprehensively characterized oxylipin and protein patterns in the human monocytic cell line THP-1 and differently polarized primary macrophages. Finally, the quantification of changes in protein and oxylipin levels induced by lipopolysaccharide stimulation and pharmaceutical treatment demonstrates its usefulness to study molecular modes of action involved in the modulation of the ARA cascade.

Functional characterization of SLC39 family members ZIP5 and ZIP10 in overexpressing HEK293 cells reveals selective copper transport activity.

Zinc is the second most prevalent metal element present in living organisms, and control of its concentration is pivotal to physiology. The amount of zinc available to the cell cytoplasm is regulated by the activity of members of the SLC39 family, the ZIP proteins. Selectivity of ZIP transporters has been the focus of earlier studies which provided a biochemical and structural basis for the selectivity for zinc over other metals such as copper, iron, and manganese. However, several previous studies have shown how certain ZIP proteins exhibit higher selectivity for metal elements other than zinc. Sequence similarities suggest an evolutionary basis for the elemental selectivity within the ZIP family. Here, by engineering HEK293 cells to overexpress ZIP proteins, we have studied the selectivity of two phylogenetic clades of ZIP proteins, that is ZIP8/ZIP14 (previously known to be iron and manganese transporters) and ZIP5/ZIP10. By incubating ZIP over-expressing cells in presence of several divalent metals, we found that ZIP5 and ZIP10 are high affinity copper transporters with greater selectivity over other elements, revealing a novel substrate signature for the ZIP5/ZIP10 clade.

Classical and quantum trial wave functions in auxiliary-field quantum Monte Carlo applied to oxygen allotropes and a CuBr2 model system.

In this work, we test a recently developed method to enhance classical auxiliary-field quantum Monte Carlo (AFQMC) calculations with quantum computers against examples from chemistry and material science, representative of classes of industry-relevant systems. As molecular test cases, we calculate the energy curve of H4 and the relative energies of ozone and singlet molecular oxygen with respect to triplet molecular oxygen, which is industrially relevant in organic oxidation reactions. We find that trial wave functions beyond single Slater determinants improve the performance of AFQMC and allow it to generate energies close to chemical accuracy compared to full configuration interaction or experimental results. In the field of material science, we study the electronic structure properties of cuprates through the quasi-1D Fermi-Hubbard model derived from CuBr2, where we find that trial wave functions with both significantly larger fidelities and lower energies over a mean-field solution do not necessarily lead to AFQMC results closer to the exact ground state energy.

Synergistic targeting of FLT3 mutations in AML via combined menin-MLL and FLT3 inhibition.

The interaction of menin (MEN1) and MLL (MLL1, KMT2A) is a dependency and provides a potential opportunity for treatment of NPM1-mutant (NPM1mut) and MLL-rearranged (MLL-r) leukemias. Concomitant activating driver mutations in the gene encoding the tyrosine kinase FLT3 occur in both leukemias and are particularly common in the NPM1mut subtype. In this study, transcriptional profiling after pharmacological inhibition of the menin-MLL complex revealed specific changes in gene expression, with downregulation of the MEIS1 transcription factor and its transcriptional target gene FLT3 being the most pronounced. Combining menin-MLL inhibition with specific small-molecule kinase inhibitors of FLT3 phosphorylation resulted in a significantly superior reduction of phosphorylated FLT3 and transcriptional suppression of genes downstream of FLT3 signaling. The drug combination induced synergistic inhibition of proliferation, as well as enhanced apoptosis, compared with single-drug treatment in models of human and murine NPM1mut and MLL-r leukemias harboring an FLT3 mutation. Primary acute myeloid leukemia (AML) cells harvested from patients with NPM1mutFLT3mut AML showed significantly better responses to combined menin and FLT3 inhibition than to single-drug or vehicle control treatment, whereas AML cells with wild-type NPM1, MLL, and FLT3 were not affected by either of the 2 drugs. In vivo treatment of leukemic animals with MLL-r FLT3mut leukemia reduced leukemia burden significantly and prolonged survival compared with results in the single-drug and vehicle control groups. Our data suggest that combined menin-MLL and FLT3 inhibition represents a novel and promising therapeutic strategy for patients with NPM1mut or MLL-r leukemia and concurrent FLT3 mutation.

Evaluation of Three Feature Dimension Reduction Techniques for Machine Learning-Based Crop Yield Prediction Models.

Machine learning (ML) has been widely used worldwide to develop crop yield forecasting models. However, it is still challenging to identify the most critical features from a dataset. Although either feature selection (FS) or feature extraction (FX) techniques have been employed, no research compares their performances and, more importantly, the benefits of combining both methods. Therefore, this paper proposes a framework that uses non-feature reduction (All-F) as a baseline to investigate the performance of FS, FX, and a combination of both (FSX). The case study employs the vegetation condition index (VCI)/temperature condition index (TCI) to develop 21 rice yield forecasting models for eight sub-regions in Vietnam based on ML methods, namely linear, support vector machine (SVM), decision tree (Tree), artificial neural network (ANN), and Ensemble. The results reveal that FSX takes full advantage of the FS and FX, leading FSX-based models to perform the best in 18 out of 21 models, while 2 (1) for FS-based (FX-based) models. These FXS-, FS-, and FX-based models improve All-F-based models at an average level of 21% and up to 60% in terms of RMSE. Furthermore, 21 of the best models are developed based on Ensemble (13 models), Tree (6 models), linear (1 model), and ANN (1 model). These findings highlight the significant role of FS, FX, and specially FSX coupled with a wide range of ML algorithms (especially Ensemble) for enhancing the accuracy of predicting crop yield.

Enhancing Crop Yield Prediction Utilizing Machine Learning on Satellite-Based Vegetation Health Indices.

Accurate crop yield forecasting is essential in the food industry's decision-making process, where vegetation condition index (VCI) and thermal condition index (TCI) coupled with machine learning (ML) algorithms play crucial roles. The drawback, however, is that a one-fits-all prediction model is often employed over an entire region without considering subregional VCI and TCI's spatial variability resulting from environmental and climatic factors. Furthermore, when using nonlinear ML, redundant VCI/TCI data present additional challenges that adversely affect the models' output. This study proposes a framework that (i) employs higher-order spatial independent component analysis (sICA), and (ii), exploits a combination of the principal component analysis (PCA) and ML (i.e., PCA-ML combination) to deal with the two challenges in order to enhance crop yield prediction accuracy. The proposed framework consolidates common VCI/TCI spatial variability into their respective subregions, using Vietnam as an example. Compared to the one-fits-all approach, subregional rice yield forecasting models over Vietnam improved by an average level of 20% up to 60%. PCA-ML combination outperformed ML-only by an average of 18.5% up to 45%. The framework generates rice yield predictions 1 to 2 months ahead of the harvest with an average of 5% error, displaying its reliability.

Next generation epigenetic modulators to target myeloid neoplasms.

PURPOSE OF REVIEW: Comprehensive sequencing studies aimed at determining the genetic landscape of myeloid neoplasms have identified epigenetic regulators to be among the most commonly mutated genes. Detailed studies have also revealed a number of epigenetic vulnerabilities. The purpose of this review is to outline these vulnerabilities and to discuss the new generation of drugs that exploit them. RECENT FINDINGS: In addition to deoxyribonucleic acid-methylation, novel epigenetic dependencies have recently been discovered in various myeloid neoplasms and many of them can be targeted pharmacologically. These include not only chromatin writers, readers, and erasers but also chromatin movers that shift nucleosomes to allow access for transcription. Inhibitors of protein-protein interactions represent a novel promising class of drugs that allow disassembly of oncogenic multiprotein complexes. SUMMARY: An improved understanding of disease-specific epigenetic vulnerabilities has led to the development of second-generation mechanism-based epigenetic drugs against myeloid neoplasms. Many of these drugs have been introduced into clinical trials and synergistic drug combination regimens have been shown to enhance efficacy and potentially prevent drug resistance.

Identification of metabolites from tandem mass spectra with a machine learning approach utilizing structural features.

MOTIVATION: Untargeted mass spectrometry (MS/MS) is a powerful method for detecting metabolites in biological samples. However, fast and accurate identification of the metabolites' structures from MS/MS spectra is still a great challenge. RESULTS: We present a new analysis method, called SubFragment-Matching (SF-Matching) that is based on the hypothesis that molecules with similar structural features will exhibit similar fragmentation patterns. We combine information on fragmentation patterns of molecules with shared substructures and then use random forest models to predict whether a given structure can yield a certain fragmentation pattern. These models can then be used to score candidate molecules for a given mass spectrum. For rapid identification, we pre-compute such scores for common biological molecular structure databases. Using benchmarking datasets, we find that our method has similar performance to CSI: FingerID and those very high accuracies can be achieved by combining our method with CSI: FingerID. Rarefaction analysis of the training dataset shows that the performance of our method will increase as more experimental data become available. AVAILABILITY AND IMPLEMENTATION: SF-Matching is available from http://www.bork.embl.de/Docu/sf_matching. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.