Identification of bacteriophage genome sequences with representation learning

Exploring deep learning in phage discovery and characterization

Bacteriophages, or bacterial viruses, play diverse ecological roles by shaping bacterial populations and also hold significant biotechnological and medical potential, including the treatment of infections caused by multidrug-resistant bacteria. The discovery of novel bacteriophages using large-scale metagenomic data has been accelerated by the accessibility of deep learning (Artificial Intelligence), the increased computing power of graphical processing units (GPUs), and new bioinformatics tools. This review addresses the recent revolution in bacteriophage research, ranging from the adoption of neural network algorithms applied to metagenomic data to the use of pre-trained language models, such as BERT, which have improved the reconstruction of viral metagenome-assembled genomes (vMAGs). This article also discusses the main aspects of bacteriophage biology using deep learning, highlighting the advances and limitations of this approach. Finally, prospects of deep-learning-based metagenomic algorithms and recommendations for future investigations are described.

Predicting viral host codon fitness and path shifting through tree-based learning on codon usage biases and genomic characteristics

Viral codon fitness (VCF) of the host and the VCF shifting has seldom been studied under quantitative measurements, although they could be concepts vital to understand pathogen epidemiology. This study demonstrates that the relative synonymous codon usage (RSCU) of virus genomes together with other genomic properties are predictive of virus host codon fitness through tree-based machine learning. Statistical analysis on the RSCU data matrix also revealed that the wobble position of the virus codons is critically important for the host codon fitness distinction. As the trained models can well characterise the host codon fitness of the viruses, the frequency and other details stored at the leaf nodes of these models can be reliably translated into human virus codon fitness score (HVCF score) as a readout of codon fitness of any virus infecting human. Specifically, we evaluated and compared HVCF of virus genome sequences from human sources and others and evaluated HVCF of SARS-CoV-2 genome sequences from NCBI virus database, where we found no obvious shifting trend in host codon fitness towards human-non-infectious. We also developed a bioinformatics tool to simulate codon-based virus fitness shifting using codon compositions of the viruses, and we found that Tylonycteris bat coronavirus HKU4 related viruses may have close relationship with SARS-CoV-2 in terms of human codon fitness. The finding of abundant synonymous mutations in the predicted codon fitness shifting path also provides new insights for evolution research and virus monitoring in environmental surveillance.

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.

PharaCon: a new framework for identifying bacteriophages via conditional representation learning

MOTIVATION

Identifying bacteriophages (phages) within metagenomic sequences is essential for understanding microbial community dynamics. Transformer-based foundation models have been successfully employed to address various biological challenges. However, these models are typically pre-trained with self-supervised tasks that do not consider label variance in the pre-training data. This presents a challenge for phage identification as pre-training on mixed bacterial and phage data may lead to information bias due to the imbalance between bacterial and phage samples.

RESULTS

To overcome this limitation, we proposed a novel conditional BERT framework that incorporates label classes as special tokens during pre-training. Specifically, our conditional BERT model attaches labels directly during tokenization, introducing label constraints into the model's input. Additionally, we introduced a new fine-tuning scheme that enables the conditional BERT to be effectively utilized for classification tasks. This framework allows the BERT model to acquire label-specific contextual representations from mixed sequence data during pre-training and applies the conditional BERT as a classifier during fine-tuning, and we named the fine-tuned model as PharaCon. We evaluated PharaCon against several existing methods on both simulated sequence datasets and real metagenomic contig datasets. The results demonstrate PharaCon's effectiveness and efficiency in phage identification, highlighting the advantages of incorporating label information during both pre-training and fine-tuning.

AVAILABILITY AND IMPLEMENTATION

The source code and associated data can be accessed at https://github.com/Celestial-Bai/PharaCon.

BFVD-a large repository of predicted viral protein structures

The AlphaFold Protein Structure Database (AFDB) is the largest repository of accurately predicted structures with taxonomic labels. Despite providing predictions for over 214 million UniProt entries, the AFDB does not cover viral sequences, severely limiting their study. To address this, we created the Big Fantastic Virus Database (BFVD), a repository of 351 242 protein structures predicted by applying ColabFold to the viral sequence representatives of the UniRef30 clusters. By utilizing homology searches across two petabases of assembled sequencing data, we improved 36% of these structure predictions beyond ColabFold's initial results. BFVD holds a unique repertoire of protein structures as over 62% of its entries show no or low structural similarity to existing repositories. We demonstrate how a substantial fraction of bacteriophage proteins, which remained unannotated based on their sequences, can be matched with similar structures from BFVD. In that, BFVD is on par with the AFDB, while holding nearly three orders of magnitude fewer structures. BFVD is an important virus-specific expansion to protein structure repositories, offering new opportunities to advance viral research. BFVD can be freely downloaded at bfvd.steineggerlab.workers.dev and queried using Foldseek and UniProt labels at bfvd.foldseek.com.

The Use of Next-Generation Sequencing in Personalized Medicine

The revolutionary progress in development of next-generation sequencing (NGS) technologies has made it possible to deliver accurate genomic information in a timely manner. Over the past several years, NGS has transformed biomedical and clinical research and found its application in the field of personalized medicine. Here we discuss the rise of personalized medicine and the history of NGS. We discuss current applications and uses of NGS in medicine, including infectious diseases, oncology, genomic medicine, and dermatology. We provide a brief discussion of selected studies where NGS was used to respond to wide variety of questions in biomedical research and clinical medicine. Finally, we discuss the challenges of implementing NGS into routine clinical use.

Advancing Phage Therapy: A Comprehensive Review of the Safety, Efficacy, and Future Prospects for the Targeted Treatment of Bacterial Infections

BACKGROUND

Phage therapy, a treatment utilizing bacteriophages to combat bacterial infections, is gaining attention as a promising alternative to antibiotics, particularly for managing antibiotic-resistant bacteria. This study aims to provide a comprehensive review of phage therapy by examining its safety, efficacy, influencing factors, future prospects, and regulatory considerations. The study also seeks to identify strategies for optimizing its application and to propose a systematic framework for its clinical implementation.

METHODS

A comprehensive analysis of preclinical studies, clinical trials, and regulatory frameworks was undertaken to evaluate the therapeutic potential of phage therapy. This included an in-depth assessment of key factors influencing clinical outcomes, such as infection site, phage-host specificity, bacterial burden, and immune response. Additionally, innovative strategies-such as combination therapies, bioengineered phages, and phage cocktails-were explored to enhance efficacy. Critical considerations related to dosing, including inoculum size, multiplicity of infection, therapeutic windows, and personalized medicine approaches, were also examined to optimize treatment outcomes.

RESULTS

Phage therapy has demonstrated a favorable safety profile in both preclinical and clinical settings, with minimal adverse effects. Its ability to specifically target harmful bacteria while preserving beneficial microbiota underpins its efficacy in treating a range of infections. However, variable outcomes in some studies highlight the importance of addressing critical factors that influence therapeutic success. Innovative approaches, including combination therapies, bioengineered phages, expanded access to diverse phage banks, phage cocktails, and personalized medicine, hold significant promise for improving efficacy. Optimizing dosing strategies remains a key area for enhancement, with critical considerations including inoculum size, multiplicity of infection, phage kinetics, resistance potential, therapeutic windows, dosing frequency, and patient-specific factors. To support the clinical application of phage therapy, a streamlined four-step guideline has been developed, providing a systematic framework for effective treatment planning and implementation.

CONCLUSION

Phage therapy offers a highly adaptable, targeted, and cost-effective approach to addressing antibiotic-resistant infections. While several critical factors must be thoroughly evaluated to optimize treatment efficacy, there remains significant potential for improvement through innovative strategies and refined methodologies. Although phage therapy has yet to achieve widespread approval in the U.S. and Europe, its accessibility through Expanded Access programs and FDA authorizations for food pathogen control underscores its promise. Established practices in countries such as Poland and Georgia further demonstrate its clinical feasibility. To enable broader adoption, regulatory harmonization and advancements in production, delivery, and quality control will be essential. Notably, the affordability and scalability of phage therapy position it as an especially valuable solution for developing regions grappling with escalating rates of antibiotic resistance.

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.

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.

VirRep: a hybrid language representation learning framework for identifying viruses from human gut metagenomes

Identifying viruses from metagenomes is a common step to explore the virus composition in the human gut. Here, we introduce VirRep, a hybrid language representation learning framework, for identifying viruses from human gut metagenomes. VirRep combines a context-aware encoder and an evolution-aware encoder to improve sequence representation by incorporating k-mer patterns and sequence homologies. Benchmarking on both simulated and real datasets with varying viral proportions demonstrates that VirRep outperforms state-of-the-art methods. When applied to fecal metagenomes from a colorectal cancer cohort, VirRep identifies 39 high-quality viral species associated with the disease, many of which cannot be detected by existing methods.

The use of next-generation sequencing in personalized medicine

The revolutionary progress in development of next-generation sequencing (NGS) technologies has made it possible to deliver accurate genomic information in a timely manner. Over the past several years, NGS has transformed biomedical and clinical research and found its application in the field of personalized medicine. Here we discuss the rise of personalized medicine and the history of NGS. We discuss current applications and uses of NGS in medicine, including infectious diseases, oncology, genomic medicine, and dermatology. We provide a brief discussion of selected studies where NGS was used to respond to wide variety of questions in biomedical research and clinical medicine. Finally, we discuss the challenges of implementing NGS into routine clinical use.

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.

Hecatomb: an integrated software platform for viral metagenomics

BACKGROUND

Modern sequencing technologies offer extraordinary opportunities for virus discovery and virome analysis. Annotation of viral sequences from metagenomic data requires a complex series of steps to ensure accurate annotation of individual reads and assembled contigs. In addition, varying study designs will require project-specific statistical analyses.

FINDINGS

Here we introduce Hecatomb, a bioinformatic platform coordinating commonly used tasks required for virome analysis. Hecatomb means "a great sacrifice." In this setting, Hecatomb is "sacrificing" false-positive viral annotations using extensive quality control and tiered-database searches. Hecatomb processes metagenomic data obtained from both short- and long-read sequencing technologies, providing annotations to individual sequences and assembled contigs. Results are provided in commonly used data formats useful for downstream analysis. Here we demonstrate the functionality of Hecatomb through the reanalysis of a primate enteric and a novel coral reef virome.

CONCLUSION

Hecatomb provides an integrated platform to manage many commonly used steps for virome characterization, including rigorous quality control, host removal, and both read- and contig-based analysis. Each step is managed using the Snakemake workflow manager with dependency management using Conda. Hecatomb outputs several tables properly formatted for immediate use within popular data analysis and visualization tools, enabling effective data interpretation for a variety of study designs. Hecatomb is hosted on GitHub (github.com/shandley/hecatomb) and is available for installation from Bioconda and PyPI.

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.

Exploring microbial functional biodiversity at the protein family level-From metagenomic sequence reads to annotated protein clusters

Metagenomics has enabled accessing the genetic repertoire of natural microbial communities. Metagenome shotgun sequencing has become the method of choice for studying and classifying microorganisms from various environments. To this end, several methods have been developed to process and analyze the sequence data from raw reads to end-products such as predicted protein sequences or families. In this article, we provide a thorough review to simplify such processes and discuss the alternative methodologies that can be followed in order to explore biodiversity at the protein family level. We provide details for analysis tools and we comment on their scalability as well as their advantages and disadvantages. Finally, we report the available data repositories and recommend various approaches for protein family annotation related to phylogenetic distribution, structure prediction and metadata enrichment.

Evaluation of computational phage detection tools for metagenomic datasets

Introduction

As new computational tools for detecting phage in metagenomes are being rapidly developed, a critical need has emerged to develop systematic benchmarks.

Methods

In this study, we surveyed 19 metagenomic phage detection tools, 9 of which could be installed and run at scale. Those 9 tools were assessed on several benchmark challenges. Fragmented reference genomes are used to assess the effects of fragment length, low viral content, phage taxonomy, robustness to eukaryotic contamination, and computational resource usage. Simulated metagenomes are used to assess the effects of sequencing and assembly quality on the tool performances. Finally, real human gut metagenomes and viromes are used to assess the differences and similarities in the phage communities predicted by the tools.

Results

We find that the various tools yield strikingly different results. Generally, tools that use a homology approach (VirSorter, MARVEL, viralVerify, VIBRANT, and VirSorter2) demonstrate low false positive rates and robustness to eukaryotic contamination. Conversely, tools that use a sequence composition approach (VirFinder, DeepVirFinder, Seeker), and MetaPhinder, have higher sensitivity, including to phages with less representation in reference databases. These differences led to widely differing predicted phage communities in human gut metagenomes, with nearly 80% of contigs being marked as phage by at least one tool and a maximum overlap of 38.8% between any two tools. While the results were more consistent among the tools on viromes, the differences in results were still significant, with a maximum overlap of 60.65%. Discussion: Importantly, the benchmark datasets developed in this study are publicly available and reusable to enable the future comparability of new tools developed.