Deep learning of a bacterial and archaeal universal language of life enables transfer learning and illuminates microbial dark matter

Revolutionizing biological digital twins: Integrating internet of bio-nano things, convolutional neural networks, and federated learning

Digital twins (DTs) are advancing biotechnology by providing digital models for drug discovery, digital health applications, and biological assets, including microorganisms. However, the hypothesis posits that implementing micro- and nanoscale DTs, especially for biological entities like bacteria, presents substantial challenges. These challenges stem from the complexities of data extraction, transmission, and computation, along with the necessity for a specialized Internet of Things (IoT) infrastructure. To address these challenges, this article proposes a novel framework that leverages bio-network technologies, including the Internet of Bio-Nano Things (IoBNT), and decentralized deep learning algorithms such as federated learning (FL) and convolutional neural networks (CNN). The methodology involves using CNNs for robust pattern recognition and FL to reduce bandwidth consumption while enhancing security. IoBNT devices are utilized for precise microscopic data acquisition and transmission, which ensures minimal error rates. The results demonstrate a multi-class classification accuracy of 98.7% across 33 bacteria categories, achieving over 99% bandwidth savings. Additionally, IoBNT integration reduces biological data transfer errors by up to 98%, even under worst-case conditions. This framework is further supported by an adaptable, user-friendly dashboard, expanding its applicability across pharmaceutical and biotechnology industries.

Genomic language models: opportunities and challenges

Large language models (LLMs) are having transformative impacts across a wide range of scientific fields, particularly in the biomedical sciences. Just as the goal of natural language processing is to understand sequences of words, a major objective in biology is to understand biological sequences. Genomic language models (gLMs), which are LLMs trained on DNA sequences, have the potential to significantly advance our understanding of genomes and how DNA elements at various scales interact to give rise to complex functions. To showcase this potential, we highlight key applications of gLMs, including functional constraint prediction, sequence design, and transfer learning. Despite notable recent progress, however, developing effective and efficient gLMs presents numerous challenges, especially for species with large, complex genomes. Here, we discuss major considerations for developing and evaluating gLMs.

FGeneBERT: function-driven pre-trained gene language model for metagenomics

Metagenomic data, comprising mixed multi-species genomes, are prevalent in diverse environments like oceans and soils, significantly impacting human health and ecological functions. However, current research relies on K-mer, which limits the capture of structurally and functionally relevant gene contexts. Moreover, these approaches struggle with encoding biologically meaningful genes and fail to address the one-to-many and many-to-one relationships inherent in metagenomic data. To overcome these challenges, we introduce FGeneBERT, a novel metagenomic pre-trained model that employs a protein-based gene representation as a context-aware and structure-relevant tokenizer. FGeneBERT incorporates masked gene modeling to enhance the understanding of inter-gene contextual relationships and triplet enhanced metagenomic contrastive learning to elucidate gene sequence-function relationships. Pre-trained on over 100 million metagenomic sequences, FGeneBERT demonstrates superior performance on metagenomic datasets at four levels, spanning gene, functional, bacterial, and environmental levels and ranging from 1 to 213 k input sequences. Case studies of ATP synthase and gene operons highlight FGeneBERT's capability for functional recognition and its biological relevance in metagenomic research.

Functional profiling of the sequence stockpile: a protein pair-based assessment of in silico prediction tools

MOTIVATION

In silico functional annotation of proteins is crucial to narrowing the sequencing-accelerated gap in our understanding of protein activities. Numerous function annotation methods exist, and their ranks have been growing, particularly so with the recent deep learning-based developments. However, it is unclear if these tools are truly predictive. As we are not aware of any methods that can identify new terms in functional ontologies, we ask if they can, at least, identify molecular functions of proteins that are non-homologous to or far-removed from known protein families.

RESULTS

Here, we explore the potential and limitations of the existing methods in predicting the molecular functions of thousands of such proteins. Lacking the "ground truth" functional annotations, we transformed the assessment of function prediction into evaluation of functional similarity of protein pairs that likely share function but are unlike any of the currently functionally annotated sequences. Notably, our approach transcends the limitations of functional annotation vocabularies, providing a means to assess different-ontology annotation methods. We find that most existing methods are limited to identifying functional similarity of homologous sequences and fail to predict the function of proteins lacking reference. Curiously, despite their seemingly unlimited by-homology scope, deep learning methods also have trouble capturing the functional signal encoded in protein sequence. We believe that our work will inspire the development of a new generation of methods that push boundaries and promote exploration and discovery in the molecular function domain.

AVAILABILITY AND IMPLEMENTATION

The data underlying this article are available at https://doi.org/10.6084/m9.figshare.c.6737127.v3. The code used to compute siblings is available openly at https://bitbucket.org/bromberglab/siblings-detector/.

RFW captures species-level metagenomic functions by integrating genome annotation information

Functional profiling of whole-metagenome shotgun sequencing (WMS) enables our understanding of microbe-host interactions. We demonstrate microbial functional information loss by current annotation methods at both the taxon and community levels, particularly at lower read depths. To address information loss, we develop a framework, RFW (reference-based functional profile inference on WMS), that utilizes information from genome functional annotations and taxonomic profiles to infer microbial function abundances from WMS. Furthermore, we provide an algorithm for absolute abundance change quantification between groups as part of the RFW framework. By applying RFW to several datasets related to autism spectrum disorder and colorectal cancer, we show that RFW augments downstream analyses, such as differential microbial function identification and association analysis between microbial function and host phenotype. RFW is open source and freely available at https://github.com/Xingyinliu-Lab/RFW.

qProtein: Exploring Physical Features of Protein Thermostability Based on Structural Proteomics

Thermostability, which is essential for the functional performance of enzymes, is largely determined by intramolecular physical interactions. Although many tools have been developed, existing computational methods have struggled to find the universal principles of protein thermostability. Recent advancements in structural proteomics have been driven by the introduction of deep neural networks such as AlphaFold2 and ESMFold. These innovations have enabled the characterization of protein structures with unprecedented speed and accuracy. Here, we introduce qProtein, a Python-implemented workflow designed for the quantitative analysis of physical interactions on the scale of structural proteomics. This platform accepts protein sequences as input and produces four structural features, including hydrophobic clusters, hydrogen bonds, electrostatic interactions, and disulfide bonds. To demonstrate the use of qProtein, we investigate the structural features related to protein thermostability in six glycoside hydrolase (GH) families, comprising a total of 3,811 protein structures. Our results indicate that in five enzyme families (GH11, GH12, GH5_2, GH10, and GH48), the thermophilic enzymes have a larger average area of hydrophobic clusters compared to the nonthermophilic enzymes within each family. Furthermore, our analysis of the local-structure regions reveals that the hydrophobic clusters are predominantly distributed in the distal regions of the GH11 enzymes. In addition, the average hydrophobic cluster area of the thermophilic enzymes is significantly higher than that of the nonthermophilic enzymes in the distal regions of the GH11 enzymes. Therefore, qProtein is a well-suited platform for analyzing the structural features of thermal stability at the level of structural proteomics. We provide the source code for qProtein at https://github.com/bj600800/qProtein, and the web server is available at http://qProtein.sdu.edu.cn:8888.

Machine learning classification of archaea and bacteria identifies novel predictive genomic features

BACKGROUND

Archaea and Bacteria are distinct domains of life that are adapted to a variety of ecological niches. Several genome-based methods have been developed for their accurate classification, yet many aspects of the specific genomic features that determine these differences are not fully understood. In this study, we used publicly available whole-genome sequences from bacteria ( N = 2546 ) and archaea ( N = 109 ). From these, a set of genomic features (nucleotide frequencies and proportions, coding sequences (CDS), non-coding, ribosomal and transfer RNA genes (ncRNA, rRNA, tRNA), Chargaff's, topological entropy and Shannon's entropy scores) was extracted and used as input data to develop machine learning models for the classification of archaea and bacteria.

RESULTS

The classification accuracy ranged from 0.993 (Random Forest) to 0.998 (Neural Networks). Over the four models, only 11 examples were misclassified, especially those belonging to the minority class (Archaea). From variable importance, tRNA topological and Shannon's entropy, nucleotide frequencies in tRNA, rRNA and ncRNA, CDS, tRNA and rRNA Chargaff's scores have emerged as the top discriminating factors. In particular, tRNA entropy (both topological and Shannon's) was the most important genomic feature for classification, pointing at the complex interactions between the genetic code, tRNAs and the translational machinery.

CONCLUSIONS

tRNA, rRNA and ncRNA genes emerged as the key genomic elements that underpin the classification of archaea and bacteria. In particular, higher nucleotide diversity was found in tRNA from bacteria compared to archaea. The analysis of the few classification errors reflects the complex phylogenetic relationships between bacteria, archaea and eukaryotes.

Navigating the archaeal frontier: insights and projections from bioinformatic pipelines

Archaea continues to be one of the least investigated domains of life, and in recent years, the advent of metagenomics has led to the discovery of many new lineages at the phylum level. For the majority, only automatic genomic annotations can provide information regarding their metabolic potential and role in the environment. Here, genomic data from 2,978 archaeal genomes was used to perform automatic annotations using bioinformatics tools, alongside synteny analysis. These automatic classifications were done to assess how good these different tools perform in relation to archaeal data. Our study revealed that even with lowered cutoffs, several functional models do not capture the recently discovered archaeal diversity. Moreover, our investigation revealed that a significant portion of archaeal genomes, approximately 42%, remain uncharacterized. In comparison, within 3,235 bacterial genomes, a diverse range of unclassified proteins is obtained, with well-studied organisms like Escherichia coli having a substantially lower proportion of uncharacterized regions, ranging from <5 to 25%, and less studied lineages being comparable to archaea with the range of 35-40% of unclassified regions. Leveraging this analysis, we were able to identify metabolic protein markers, thereby providing insights into the metabolism of the archaea in our dataset. Our findings underscore a substantial gap between automatic classification tools and the comprehensive mapping of archaeal metabolism. Despite advances in computational approaches, a significant portion of archaeal genomes remains unexplored, highlighting the need for extensive experimental validation in this domain, as well as more refined annotation methods. This study contributes to a better understanding of archaeal metabolism and underscores the importance of further research in elucidating the functional potential of archaeal genomes.

Genomic Language Models: Opportunities and Challenges

Large language models (LLMs) are having transformative impacts across a wide range of scientific fields, particularly in the biomedical sciences. Just as the goal of Natural Language Processing is to understand sequences of words, a major objective in biology is to understand biological sequences. Genomic Language Models (gLMs), which are LLMs trained on DNA sequences, have the potential to significantly advance our understanding of genomes and how DNA elements at various scales interact to give rise to complex functions. To showcase this potential, we highlight key applications of gLMs, including functional constraint prediction, sequence design, and transfer learning. Despite notable recent progress, however, developing effective and efficient gLMs presents numerous challenges, especially for species with large, complex genomes. Here, we discuss major considerations for developing and evaluating gLMs.

DeepMineLys: Deep mining of phage lysins from human microbiome

Vast shotgun metagenomics data remain an underutilized resource for novel enzymes. Artificial intelligence (AI) has increasingly been applied to protein mining, but its conventional performance evaluation is interpolative in nature, and these trained models often struggle to extrapolate effectively when challenged with unknown data. In this study, we present a framework (DeepMineLys [deep mining of phage lysins from human microbiome]) based on the convolutional neural network (CNN) to identify phage lysins from three human microbiome datasets. When validated with an independent dataset, our method achieved an F1-score of 84.00%, surpassing existing methods by 20.84%. We expressed 16 lysin candidates from the top 100 sequences in E. coli, confirming 11 as active. The best one displayed an activity 6.2-fold that of lysozyme derived from hen egg white, establishing it as the most potent lysin from the human microbiome. Our study also underscores several important issues when applying AI to biology questions. This framework should be applicable for mining other proteins.

A foundational large language model for edible plant genomes

Significant progress has been made in the field of plant genomics, as demonstrated by the increased use of high-throughput methodologies that enable the characterization of multiple genome-wide molecular phenotypes. These findings have provided valuable insights into plant traits and their underlying genetic mechanisms, particularly in model plant species. Nonetheless, effectively leveraging them to make accurate predictions represents a critical step in crop genomic improvement. We present AgroNT, a foundational large language model trained on genomes from 48 plant species with a predominant focus on crop species. We show that AgroNT can obtain state-of-the-art predictions for regulatory annotations, promoter/terminator strength, tissue-specific gene expression, and prioritize functional variants. We conduct a large-scale in silico saturation mutagenesis analysis on cassava to evaluate the regulatory impact of over 10 million mutations and provide their predicted effects as a resource for variant characterization. Finally, we propose the use of the diverse datasets compiled here as the Plants Genomic Benchmark (PGB), providing a comprehensive benchmark for deep learning-based methods in plant genomic research. The pre-trained AgroNT model is publicly available on HuggingFace at https://huggingface.co/InstaDeepAI/agro-nucleotide-transformer-1b  for future research purposes.

Unravelling biosynthesis and biodegradation potentials of microbial dark matters in hypersaline lakes

Biosynthesis and biodegradation of microorganisms critically underpin the development of biotechnology, new drugs and therapies, and environmental remediation. However, most uncultured microbial species along with their metabolic capacities in extreme environments, remain obscured. Here we unravel the metabolic potential of microbial dark matters (MDMs) in four deep-inland hypersaline lakes in Xinjiang, China. Utilizing metagenomic binning, we uncovered a rich diversity of 3030 metagenome-assembled genomes (MAGs) across 82 phyla, revealing a substantial portion, 2363 MAGs, as previously unclassified at the genus level. These unknown MAGs displayed unique distribution patterns across different lakes, indicating a strong correlation with varied physicochemical conditions. Our analysis revealed an extensive array of 9635 biosynthesis gene clusters (BGCs), with a remarkable 9403 being novel, suggesting untapped biotechnological potential. Notably, some MAGs from potentially new phyla exhibited a high density of these BGCs. Beyond biosynthesis, our study also identified novel biodegradation pathways, including dehalogenation, anaerobic ammonium oxidation (Anammox), and degradation of polycyclic aromatic hydrocarbons (PAHs) and plastics, in previously unknown microbial clades. These findings significantly enrich our understanding of biosynthesis and biodegradation processes and open new avenues for biotechnological innovation, emphasizing the untapped potential of microbial diversity in hypersaline environments.

SAFPred: synteny-aware gene function prediction for bacteria using protein embeddings

MOTIVATION

Today, we know the function of only a small fraction of the protein sequences predicted from genomic data. This problem is even more salient for bacteria, which represent some of the most phylogenetically and metabolically diverse taxa on Earth. This low rate of bacterial gene annotation is compounded by the fact that most function prediction algorithms have focused on eukaryotes, and conventional annotation approaches rely on the presence of similar sequences in existing databases. However, often there are no such sequences for novel bacterial proteins. Thus, we need improved gene function prediction methods tailored for bacteria. Recently, transformer-based language models-adopted from the natural language processing field-have been used to obtain new representations of proteins, to replace amino acid sequences. These representations, referred to as protein embeddings, have shown promise for improving annotation of eukaryotes, but there have been only limited applications on bacterial genomes.

RESULTS

To predict gene functions in bacteria, we developed SAFPred, a novel synteny-aware gene function prediction tool based on protein embeddings from state-of-the-art protein language models. SAFpred also leverages the unique operon structure of bacteria through conserved synteny. SAFPred outperformed both conventional sequence-based annotation methods and state-of-the-art methods on multiple bacterial species, including for distant homolog detection, where the sequence similarity to the proteins in the training set was as low as 40%. Using SAFPred to identify gene functions across diverse enterococci, of which some species are major clinical threats, we identified 11 previously unrecognized putative novel toxins, with potential significance to human and animal health.

AVAILABILITY AND IMPLEMENTATION

https://github.com/AbeelLab/safpred.

Single cell analysis reveals the roles and regulatory mechanisms of type-I interferons in Parkinson's disease

The pathogenesis of Parkinson's disease (PD) is strongly associated with neuroinflammation, and type I interferons (IFN-I) play a crucial role in regulating immune and inflammatory responses. However, the specific features of IFN in different cell types and the underlying mechanisms of PD have yet to be fully described. In this study, we analyzed the GSE157783 dataset, which includes 39,024 single-cell RNA sequencing results for five PD patients and six healthy controls from the Gene Expression Omnibus database. After cell type annotation, we intersected differentially expressed genes in each cell subcluster with genes collected in The Interferome database to generate an IFN-I-stimulated gene set (ISGs). Based on this gene set, we used the R package AUCell to score each cell, representing the IFN-I activity. Additionally, we performed monocle trajectory analysis, and single-cell regulatory network inference and clustering (SCENIC) to uncover the underlying mechanisms. In silico gene perturbation and subsequent experiments confirm NFATc2 regulation of type I interferon response and neuroinflammation. Our analysis revealed that microglia, endothelial cells, and pericytes exhibited the highest activity of IFN-I. Furthermore, single-cell trajectory detection demonstrated that microglia in the midbrain of PD patients were in a pro-inflammatory activation state, which was validated in the 1-Methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)-induced PD mouse model as well. We identified transcription factors NFATc2, which was significantly up-regulated and involved in the expression of ISGs and activation of microglia in PD. In the 1-Methyl-4-phenylpyridinium (MPP+)-induced BV2 cell model, the suppression of NFATc2 resulted in a reduction in IFN-β levels, impeding the phosphorylation of STAT1, and attenuating the activation of the NF-κB pathway. Furthermore, the downregulation of NFATc2 mitigated the detrimental effects on SH-SY5Y cells co-cultured in conditioned medium. Our study highlights the critical role of microglia in type I interferon responses in PD. Additionally, we identified transcription factors NFATc2 as key regulators of aberrant type I interferon responses and microglial pro-inflammatory activation in PD. These findings provide new insights into the pathogenesis of PD and may have implications for the development of novel therapeutic strategies.

Self-supervised learning on millions of primary RNA sequences from 72 vertebrates improves sequence-based RNA splicing prediction

Language models pretrained by self-supervised learning (SSL) have been widely utilized to study protein sequences, while few models were developed for genomic sequences and were limited to single species. Due to the lack of genomes from different species, these models cannot effectively leverage evolutionary information. In this study, we have developed SpliceBERT, a language model pretrained on primary ribonucleic acids (RNA) sequences from 72 vertebrates by masked language modeling, and applied it to sequence-based modeling of RNA splicing. Pretraining SpliceBERT on diverse species enables effective identification of evolutionarily conserved elements. Meanwhile, the learned hidden states and attention weights can characterize the biological properties of splice sites. As a result, SpliceBERT was shown effective on several downstream tasks: zero-shot prediction of variant effects on splicing, prediction of branchpoints in humans, and cross-species prediction of splice sites. Our study highlighted the importance of pretraining genomic language models on a diverse range of species and suggested that SSL is a promising approach to enhance our understanding of the regulatory logic underlying genomic sequences.

ProkBERT family: genomic language models for microbiome applications

Background

In the evolving landscape of microbiology and microbiome analysis, the integration of machine learning is crucial for understanding complex microbial interactions, and predicting and recognizing novel functionalities within extensive datasets. However, the effectiveness of these methods in microbiology faces challenges due to the complex and heterogeneous nature of microbial data, further complicated by low signal-to-noise ratios, context-dependency, and a significant shortage of appropriately labeled datasets. This study introduces the ProkBERT model family, a collection of large language models, designed for genomic tasks. It provides a generalizable sequence representation for nucleotide sequences, learned from unlabeled genome data. This approach helps overcome the above-mentioned limitations in the field, thereby improving our understanding of microbial ecosystems and their impact on health and disease.

Methods

ProkBERT models are based on transfer learning and self-supervised methodologies, enabling them to use the abundant yet complex microbial data effectively. The introduction of the novel Local Context-Aware (LCA) tokenization technique marks a significant advancement, allowing ProkBERT to overcome the contextual limitations of traditional transformer models. This methodology not only retains rich local context but also demonstrates remarkable adaptability across various bioinformatics tasks.

Results

In practical applications such as promoter prediction and phage identification, the ProkBERT models show superior performance. For promoter prediction tasks, the top-performing model achieved a Matthews Correlation Coefficient (MCC) of 0.74 for E. coli and 0.62 in mixed-species contexts. In phage identification, ProkBERT models consistently outperformed established tools like VirSorter2 and DeepVirFinder, achieving an MCC of 0.85. These results underscore the models' exceptional accuracy and generalizability in both supervised and unsupervised tasks.

Conclusions

The ProkBERT model family is a compact yet powerful tool in the field of microbiology and bioinformatics. Its capacity for rapid, accurate analyses and its adaptability across a spectrum of tasks marks a significant advancement in machine learning applications in microbiology. The models are available on GitHub (https://github.com/nbrg-ppcu/prokbert) and HuggingFace (https://huggingface.co/nerualbioinfo) providing an accessible tool for the community.

The energetics and evolution of oxidoreductases in deep time

The core metabolic reactions of life drive electrons through a class of redox protein enzymes, the oxidoreductases. The energetics of electron flow is determined by the redox potentials of organic and inorganic cofactors as tuned by the protein environment. Understanding how protein structure affects oxidation-reduction energetics is crucial for studying metabolism, creating bioelectronic systems, and tracing the history of biological energy utilization on Earth. We constructed ProtReDox (https://protein-redox-potential.web.app), a manually curated database of experimentally determined redox potentials. With over 500 measurements, we can begin to identify how proteins modulate oxidation-reduction energetics across the tree of life. By mapping redox potentials onto networks of oxidoreductase fold evolution, we can infer the evolution of electron transfer energetics over deep time. ProtReDox is designed to include user-contributed submissions with the intention of making it a valuable resource for researchers in this field.

r/K selection of GC content in prokaryotes

The guanine/cytosine (GC) content of prokaryotic genomes is species-specific, taking values from 16% to 77%. This diversity of selection for GC content remains contentious. We analyse the correlations between GC content and a range of phenotypic and genotypic data in thousands of prokaryotes. GC content integrates well with these traits into r/K selection theory when phenotypic plasticity is considered. High GC-content prokaryotes are r-strategists with cheaper descendants thanks to a lower average amino acid metabolic cost, colonize unstable environments thanks to flagella and a bacillus form and are generalists in terms of resource opportunism and their defence mechanisms. Low GC content prokaryotes are K-strategists specialized for stable environments that maintain homeostasis via a high-cost outer cell membrane and endospore formation as a response to nutrient deprivation, and attain a higher nutrient-to-biomass yield. The lower proteome cost of high GC content prokaryotes is driven by the association between GC-rich codons and cheaper amino acids in the genetic code, while the correlation between GC content and genome size may be partly due to functional diversity driven by r/K selection. In all, molecular diversity in the GC content of prokaryotes may be a consequence of ecological r/K selection.

A toolbox of machine learning software to support microbiome analysis

The human microbiome has become an area of intense research due to its potential impact on human health. However, the analysis and interpretation of this data have proven to be challenging due to its complexity and high dimensionality. Machine learning (ML) algorithms can process vast amounts of data to uncover informative patterns and relationships within the data, even with limited prior knowledge. Therefore, there has been a rapid growth in the development of software specifically designed for the analysis and interpretation of microbiome data using ML techniques. These software incorporate a wide range of ML algorithms for clustering, classification, regression, or feature selection, to identify microbial patterns and relationships within the data and generate predictive models. This rapid development with a constant need for new developments and integration of new features require efforts into compile, catalog and classify these tools to create infrastructures and services with easy, transparent, and trustable standards. Here we review the state-of-the-art for ML tools applied in human microbiome studies, performed as part of the COST Action ML4Microbiome activities. This scoping review focuses on ML based software and framework resources currently available for the analysis of microbiome data in humans. The aim is to support microbiologists and biomedical scientists to go deeper into specialized resources that integrate ML techniques and facilitate future benchmarking to create standards for the analysis of microbiome data. The software resources are organized based on the type of analysis they were developed for and the ML techniques they implement. A description of each software with examples of usage is provided including comments about pitfalls and lacks in the usage of software based on ML methods in relation to microbiome data that need to be considered by developers and users. This review represents an extensive compilation to date, offering valuable insights and guidance for researchers interested in leveraging ML approaches for microbiome analysis.

A genomic catalogue of soil microbiomes boosts mining of biodiversity and genetic resources

Soil harbors a vast expanse of unidentified microbes, termed as microbial dark matter, presenting an untapped reservo)ir of microbial biodiversity and genetic resources, but has yet to be fully explored. In this study, we conduct a large-scale excavation of soil microbial dark matter by reconstructing 40,039 metagenome-assembled genome bins (the SMAG catalogue) from 3304 soil metagenomes. We identify 16,530 of 21,077 species-level genome bins (SGBs) as unknown SGBs (uSGBs), which expand archaeal and bacterial diversity across the tree of life. We also illustrate the pivotal role of uSGBs in augmenting soil microbiome's functional landscape and intra-species genome diversity, providing large proportions of the 43,169 biosynthetic gene clusters and 8545 CRISPR-Cas genes. Additionally, we determine that uSGBs contributed 84.6% of previously unexplored viral-host associations from the SMAG catalogue. The SMAG catalogue provides an useful genomic resource for further studies investigating soil microbial biodiversity and genetic resources.

DNA language models are powerful predictors of genome-wide variant effects

The expanding catalog of genome-wide association studies (GWAS) provides biological insights across a variety of species, but identifying the causal variants behind these associations remains a significant challenge. Experimental validation is both labor-intensive and costly, highlighting the need for accurate, scalable computational methods to predict the effects of genetic variants across the entire genome. Inspired by recent progress in natural language processing, unsupervised pretraining on large protein sequence databases has proven successful in extracting complex information related to proteins. These models showcase their ability to learn variant effects in coding regions using an unsupervised approach. Expanding on this idea, we here introduce the Genomic Pre-trained Network (GPN), a model designed to learn genome-wide variant effects through unsupervised pretraining on genomic DNA sequences. Our model also successfully learns gene structure and DNA motifs without any supervision. To demonstrate its utility, we train GPN on unaligned reference genomes of Arabidopsis thaliana and seven related species within the Brassicales order and evaluate its ability to predict the functional impact of genetic variants in A. thaliana by utilizing allele frequencies from the 1001 Genomes Project and a comprehensive database of GWAS. Notably, GPN outperforms predictors based on popular conservation scores such as phyloP and phastCons. Our predictions for A. thaliana can be visualized as sequence logos in the UCSC Genome Browser (https://genome.ucsc.edu/s/gbenegas/gpn-arabidopsis). We provide code (https://github.com/songlab-cal/gpn) to train GPN for any given species using its DNA sequence alone, enabling unsupervised prediction of variant effects across the entire genome.

Learning from the unknown: exploring the range of bacterial functionality

Determining the repertoire of a microbe's molecular functions is a central question in microbial biology. Modern techniques achieve this goal by comparing microbial genetic material against reference databases of functionally annotated genes/proteins or known taxonomic markers such as 16S rRNA. Here, we describe a novel approach to exploring bacterial functional repertoires without reference databases. Our Fusion scheme establishes functional relationships between bacteria and assigns organisms to Fusion-taxa that differ from otherwise defined taxonomic clades. Three key findings of our work stand out. First, bacterial functional comparisons outperform marker genes in assigning taxonomic clades. Fusion profiles are also better for this task than other functional annotation schemes. Second, Fusion-taxa are robust to addition of novel organisms and are, arguably, able to capture the environment-driven bacterial diversity. Finally, our alignment-free nucleic acid-based Siamese Neural Network model, created using Fusion functions, enables finding shared functionality of very distant, possibly structurally different, microbial homologs. Our work can thus help annotate functional repertoires of bacterial organisms and further guide our understanding of microbial communities.

A Transiently Hypersaline Microbial Mat Harbors a Diverse and Stable Archaeal Community in the Cuatro Cienegas Basin, Mexico

Microbial mats are biologically diverse communities that are analogs to some of the earliest ecosystems on Earth. In this study, we describe a unique transiently hypersaline microbial mat uncovered in a shallow pond within the Cuatro Cienegas Basin (CCB) in northern México. The CCB is an endemism-rich site that harbors living stromatolites that have been studied to understand the conditions of the Precambrian Earth. These microbial mats form elastic domes filled with biogenic gas, and the mats have a relatively large and stable subpopulation of archaea. For this reason, this site has been termed archaean domes (AD). The AD microbial community was analyzed by metagenomics over three seasons. The mat exhibited a highly diverse prokaryotic community dominated by bacteria. Bacterial sequences are represented in 37 phyla, mainly Proteobacteria, Firmicutes, and Actinobacteria, that together comprised >50% of the sequences from the mat. Archaea represented up to 5% of the retrieved sequences, with up to 230 different archaeal species that belong to 5 phyla (Euryarchaeota, Crenarchaeota, Thaumarchaeota, Korarchaeota, and Nanoarchaeota). The archaeal taxa showed low variation despite fluctuations in water and nutrient availability. In addition, predicted functions highlight stress responses to extreme conditions present in the AD, including salinity, pH, and water/drought fluctuation. The observed complexity of the AD mat thriving in high pH and fluctuating water and salt conditions within the CCB provides an extant model of great value for evolutionary studies, as well as a suitable analog to the early Earth and Mars.

Deep learning to estimate lithium-ion battery state of health without additional degradation experiments

State of health is a critical state which evaluates the degradation level of batteries. However, it cannot be measured directly but requires estimation. While accurate state of health estimation has progressed markedly, the time- and resource-consuming degradation experiments to generate target battery labels hinder the development of state of health estimation methods. In this article, we design a deep-learning framework to enable the estimation of battery state of health in the absence of target battery labels. This framework integrates a swarm of deep neural networks equipped with domain adaptation to produce accurate estimation. We employ 65 commercial batteries from 5 different manufacturers to generate 71,588 samples for cross-validation. The validation results indicate that the proposed framework can ensure absolute errors of less than 3% for 89.4% of samples (less than 5% for 98.9% of samples), with a maximum absolute error of less than 8.87% in the absence of target labels. This work emphasizes the power of deep learning in precluding degradation experiments and highlights the promise of rapid development of battery management algorithms for new-generation batteries using only previous experimental data.

'Small Data' for big insights in ecology

Big Data science has significantly furthered our understanding of complex systems by harnessing large volumes of data, generated at high velocity and in great variety. However, there is a risk that Big Data collection is prioritised to the detriment of 'Small Data' (data with few observations). This poses a particular risk to ecology where Small Data abounds. Machine learning experts are increasingly looking to Small Data to drive the next generation of innovation, leading to development in methods for Small Data such as transfer learning, knowledge graphs, and synthetic data. Meanwhile, meta-analysis and causal reasoning approaches are evolving to provide new insights from Small Data. These advances should add value to high-quality Small Data catalysing future insights for ecology.

Meta-Viromic Sequencing Reveals Virome Characteristics of Mosquitoes and Culicoides on Zhoushan Island, China

Mosquitoes and biting Culicoides species are arbovirus vectors. Effective virome profile surveillance is essential for the prevention and control of insect-borne diseases. From June to September 2021, we collected eight species of female mosquito and Culicoides on Zhoushan Island, China, and used meta-viromic sequencing to analyze their virome compositions and characteristics. The classified virus reads were distributed in 191 genera in 66 families. The virus sequences in mosquitoes with the largest proportions were Iflaviridae (30.03%), Phasmaviridae (23.09%), Xinmoviridae (21.82%), Flaviviridae (13.44%), and Rhabdoviridae (8.40%). Single-strand RNA⁺ viruses formed the largest proportions of viruses in all samples. Blood meals indicated that blood-sucking mosquito hosts were mainly chicken, duck, pig, and human, broadly consistent with the habitats where the mosquitoes were collected. Novel viruses of the Orthobunyavirus, Narnavirus, and Iflavirus genera were found in Culicoides by de-novo assembly. The viruses with vertebrate hosts carried by mosquitoes and Culicoides also varied widely. The analysis of unclassified viruses and deep-learning analysis of the "dark matter" in the meta-viromic sequencing data revealed the presence of a large number of unknown viruses. IMPORTANCE The monitoring of the viromes of mosquitoes and Culicoides, widely distributed arbovirus transmission vectors, is crucial to evaluate the risk of infectious disease transmission. In this study, the compositions of the viromes of mosquitoes and Culicoides on Zhoushan Island varied widely and were related mainly to the host species, with different host species having different core viromes. and many unknown sequences in the Culicoides viromes remain to be annotated, suggesting the presence of a large number of unknown viruses.