The advent of rapid whole-genome sequencing has created new opportunities for computational prediction of antimicrobial resistance (AMR) phenotypes from genomic data. Both rule-based and machine learning (ML) approaches have been explored for this task, but systematic benchmarking is still needed. Here, we evaluated four state-of-the-art ML methods (Kover, PhenotypeSeeker, Seq2Geno2Pheno and Aytan-Aktug), an ML baseline and the rule-based ResFinder by training and testing each of them across 78 species-antibiotic datasets, using a rigorous benchmarking workflow that integrates three evaluation approaches, each paired with three distinct sample splitting methods. Our analysis revealed considerable variation in the performance across techniques and datasets. Whereas ML methods generally excelled for closely related strains, ResFinder excelled for handling divergent genomes. Overall, Kover most frequently ranked top among the ML approaches, followed by PhenotypeSeeker and Seq2Geno2Pheno. AMR phenotypes for antibiotic classes such as macrolides and sulfonamides were predicted with the highest accuracies. The quality of predictions varied substantially across species-antibiotic combinations, particularly for beta-lactams; across species, resistance phenotyping of the beta-lactams compound, aztreonam, amoxicillin/clavulanic acid, cefoxitin, ceftazidime and piperacillin/tazobactam, alongside tetracyclines demonstrated more variable performance than the other benchmarked antibiotics. By organism, Campylobacter jejuni and Enterococcus faecium phenotypes were more robustly predicted than those of Escherichia coli, Staphylococcus aureus, Salmonella enterica, Neisseria gonorrhoeae, Klebsiella pneumoniae, Pseudomonas aeruginosa, Acinetobacter baumannii, Streptococcus pneumoniae and Mycobacterium tuberculosis. In addition, our study provides software recommendations for each species-antibiotic combination. It furthermore highlights the need for optimization for robust clinical applications, particularly for strains that diverge substantially from those used for training.

Deep learning has enabled outstanding progress on bioinformatics datasets and a variety of tasks, such as protein structure prediction, identification of regulatory regions, genome annotation, and interpretation of the noncoding genome. The layout and configuration of neural networks used for these tasks have mostly been developed manually by human experts, which is a time-consuming and error-prone process. Therefore, there is growing interest in automated neural architecture search (NAS) methods in bioinformatics. In this paper, we present a novel search space for NAS algorithms that operate on genome data, thus creating extensions for existing NAS algorithms for sequence data that we name Genome-DARTS, Genome-P-DARTS, Genome-BONAS, Genome-SH, and Genome-RS. Moreover, we introduce two novel NAS algorithms, CWP-DARTS and EDPDARTS, that build on and extend the idea of P-DARTS. We evaluate the presented methods and compare them to manually designed neural architectures on a widely used genome sequence machine learning task to show that NAS methods can be adapted well for bioinformatics sequence datasets. Our experiments show that architectures optimized by our NAS methods outperform manually developed architectures while having significantly fewer parameters.

MOTIVATION: Gene annotation is the problem of mapping proteins to their functions represented as Gene Ontology (GO) terms, typically inferred based on the primary sequences. Gene annotation is a multi-label multi-class classification problem, which has generated growing interest for its uses in the characterization of millions of proteins with unknown functions. However, there is no standard GO dataset used for benchmarking the newly developed new machine learning models within the bioinformatics community. Thus, the significance of improvements for these models remains unclear. RESULTS: The Gene Benchmarking database is the first effort to provide an easy-to-use and configurable hub for the learning and evaluation of gene annotation models. It provides easy access to pre-specified datasets and takes the non-trivial steps of preprocessing and filtering all data according to custom presets using a web interface. The GO bench web application can also be used to evaluate and display any trained model on leaderboards for annotation tasks. AVAILABILITY AND IMPLEMENTATION: The GO Benchmarking dataset is freely available at www.gobench.org. Code is hosted at github.com/mofradlab, with repositories for website code, core utilities and examples of usage (Supplementary Section S.7). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Single-cell genome sequencing provides a highly granular view of biological systems but is affected by high error rates, allelic amplification bias, and uneven genome coverage. This creates a need for data-specific computational methods, for purposes such as for cell lineage tree inference. The objective of cell lineage tree reconstruction is to infer the evolutionary process that generated a set of observed cell genomes. Lineage trees may enable a better understanding of tumor formation and growth, as well as of organ development for healthy body cells. We describe a method, Scelestial, for lineage tree reconstruction from single-cell data, which is based on an approximation algorithm for the Steiner tree problem and is a generalization of the neighbor-joining method. We adapt the algorithm to efficiently select a limited subset of potential sequences as internal nodes, in the presence of missing values, and to minimize cost by lineage tree-based missing value imputation. In a comparison against seven state-of-the-art single-cell lineage tree reconstruction algorithms-BitPhylogeny, OncoNEM, SCITE, SiFit, SASC, SCIPhI, and SiCloneFit-on simulated and real single-cell tumor samples, Scelestial performed best at reconstructing trees in terms of accuracy and run time. Scelestial has been implemented in C++. It is also available as an R package named RScelestial.

Pretrained representations have recently gained attention in various machine learning applications. Nonetheless, the high computational costs associated with training these models have motivated alternative approaches for representation learning. Herein we introduce TripletProt, a new approach for protein representation learning based on the Siamese neural networks. Representation learning of biological entities which capture essential features can alleviate many of the challenges associated with supervised learning in bioinformatics. The most important distinction of our proposed method is relying on the protein-protein interaction (PPI) network. The computational cost of the generated representations for any potential application is significantly lower than comparable methods since the length of the representations is significantly smaller than that in other approaches. TripletProt offers great potentials for the protein informatics tasks and can be widely applied to similar tasks. We evaluate TripletProt comprehensively in protein functional annotation tasks including sub-cellular localization (14 categories) and gene ontology prediction (more than 2000 classes), which are both challenging multi-class, multi-label classification machine learning problems. We compare the performance of TripletProt with the state-of-the-art approaches including a recurrent language model-based approach (i.e., UniRep), as well as a protein-protein interaction (PPI) network and sequence-based method (i.e., DeepGO). Our TripletProt showed an overall improvement of F1 score in the above mentioned comprehensive functional annotation tasks, solely relying on the PPI network. Availability: The source code and datasets are available at https://github.com/EsmaeilNourani/TripletProt.

Infection with human cytomegalovirus (HCMV) can cause severe complications in immunocompromised individuals and congenitally infected children. Characterizing heterogeneous viral populations and their evolution by high-throughput sequencing of clinical specimens requires the accurate assembly of individual strains or sequence variants and suitable variant calling methods. However, the performance of most methods has not been assessed for populations composed of low divergent viral strains with large genomes, such as HCMV. In an extensive benchmarking study, we evaluated 15 assemblers and 6 variant callers on 10 lab-generated benchmark data sets created with two different library preparation protocols, to identify best practices and challenges for analyzing such data. Most assemblers, especially metaSPAdes and IVA, performed well across a range of metrics in recovering abundant strains. However, only one, Savage, recovered low abundant strains and in a highly fragmented manner. Two variant callers, LoFreq and VarScan2, excelled across all strain abundances. Both shared a large fraction of false positive variant calls, which were strongly enriched in T to G changes in a 'G.G' context. The magnitude of this context-dependent systematic error is linked to the experimental protocol. We provide all benchmarking data, results and the entire benchmarking workflow named QuasiModo, Quasispecies Metric determination on omics, under the GNU General Public License v3.0 (https://github.com/hzi-bifo/Quasimodo), to enable full reproducibility and further benchmarking on these and other data.

SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) is a novel virus of the family Coronaviridae. The virus causes the infectious disease COVID-19. The biology of coronaviruses has been studied for many years. However, bioinformatics tools designed explicitly for SARS-CoV-2 have only recently been developed as a rapid reaction to the need for fast detection, understanding and treatment of COVID-19. To control the ongoing COVID-19 pandemic, it is of utmost importance to get insight into the evolution and pathogenesis of the virus. In this review, we cover bioinformatics workflows and tools for the routine detection of SARS-CoV-2 infection, the reliable analysis of sequencing data, the tracking of the COVID-19 pandemic and evaluation of containment measures, the study of coronavirus evolution, the discovery of potential drug targets and development of therapeutic strategies. For each tool, we briefly describe its use case and how it advances research specifically for SARS-CoV-2. All tools are free to use and available online, either through web applications or public code repositories. Contact:[email protected].

MOTIVATION: B-cell epitopes (BCEs) play a pivotal role in the development of peptide vaccines, immuno-diagnostic reagents and antibody production, and thus in infectious disease prevention and diagnostics in general. Experimental methods used to determine BCEs are costly and time-consuming. Therefore, it is essential to develop computational methods for the rapid identification of BCEs. Although several computational methods have been developed for this task, generalizability is still a major concern, where cross-testing of the classifiers trained and tested on different datasets has revealed accuracies of 51-53%. RESULTS: We describe a new method called EpitopeVec, which uses a combination of residue properties, modified antigenicity scales, and protein language model-based representations (protein vectors) as features of peptides for linear BCE predictions. Extensive benchmarking of EpitopeVec and other state-of-the-art methods for linear BCE prediction on several large and small datasets, as well as cross-testing, demonstrated an improvement in the performance of EpitopeVec over other methods in terms of accuracy and area under the curve. As the predictive performance depended on the species origin of the respective antigens (viral, bacterial and eukaryotic), we also trained our method on a large viral dataset to create a dedicated linear viral BCE predictor with improved cross-testing performance. AVAILABILITY AND IMPLEMENTATION: The software is available at https://github.com/hzi-bifo/epitope-prediction. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Influenza A viruses cause seasonal epidemics and occasional pandemics in the human population. While the worldwide circulation of seasonal influenza is at least partly understood, the exact migration patterns between countries, states or cities are not well studied. Here, we use the Sankoff algorithm for parsimonious phylogeographic reconstruction together with effective distances based on a worldwide air transportation network. By first simulating geographic spread and then phylogenetic trees and genetic sequences, we confirmed that reconstructions with effective distances inferred phylogeographic spread more accurately than reconstructions with geographic distances and Bayesian reconstructions with BEAST that do not use any distance information, and led to comparable results to the Bayesian reconstruction using distance information via a generalized linear model. Our method extends Bayesian methods that estimate rates from the data by using fine-grained locations like airports and inferring intermediate locations not observed among sampled isolates. When applied to sequence data of the pandemic H1N1 influenza A virus in 2009, our approach correctly inferred the origin and proposed airports mainly involved in the spread of the virus. In case of a novel outbreak, this approach allows to rapidly analyze sequence data and infer origin and spread routes to improve disease surveillance and control.

Bioinformatics, bty296, https://doi.org/10.1093/bioinformatics/bty296 The author wishes to inform readers that the affiliations for the authors were incorrect in the original paper. They appear correctly above. The paper has been corrected online.

SUMMARY: Identifying distinctive taxa for micro-biome-related diseases is considered key to the establishment of diagnosis and therapy options in precision medicine and imposes high demands on the accuracy of micro-biome analysis techniques. We propose an alignment- and reference- free subsequence based 16S rRNA data analysis, as a new paradigm for micro-biome phenotype and biomarker detection. Our method, called DiTaxa, substitutes standard operational taxonomic unit (OTU)-clustering by segmenting 16S rRNA reads into the most frequent variable-length subsequences. We compared the performance of DiTaxa to the state-of-the-art methods in phenotype and biomarker detection, using human-associated 16S rRNA samples for periodontal disease, rheumatoid arthritis and inflammatory bowel diseases, as well as a synthetic benchmark dataset. DiTaxa performed competitively to the k-mer based state-of-the-art approach in phenotype prediction while outperforming the OTU-based state-of-the-art approach in finding biomarkers in both resolution and coverage evaluated over known links from literature and synthetic benchmark datasets. AVAILABILITY AND IMPLEMENTATION: DiTaxa is available under the Apache 2 license at http://llp.berkeley.edu/ditaxa. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Motivation: Microbial communities play important roles in the function and maintenance of various biosystems, ranging from the human body to the environment. A major challenge in microbiome research is the classification of microbial communities of different environments or host phenotypes. The most common and cost-effective approach for such studies to date is 16S rRNA gene sequencing. Recent falls in sequencing costs have increased the demand for simple, efficient and accurate methods for rapid detection or diagnosis with proved applications in medicine, agriculture and forensic science. We describe a reference- and alignment-free approach for predicting environments and host phenotypes from 16S rRNA gene sequencing based on k-mer representations that benefits from a bootstrapping framework for investigating the sufficiency of shallow sub-samples. Deep learning methods as well as classical approaches were explored for predicting environments and host phenotypes. Results: A k-mer distribution of shallow sub-samples outperformed Operational Taxonomic Unit (OTU) features in the tasks of body-site identification and Crohn's disease prediction. Aside from being more accurate, using k-mer features in shallow sub-samples allows (i) skipping computationally costly sequence alignments required in OTU-picking and (ii) provided a proof of concept for the sufficiency of shallow and short-length 16S rRNA sequencing for phenotype prediction. In addition, k-mer features predicted representative 16S rRNA gene sequences of 18 ecological environments, and 5 organismal environments with high macro-F1 scores of 0.88 and 0.87. For large datasets, deep learning outperformed classical methods such as Random Forest and Support Vector Machine. Availability and implementation: The software and datasets are available at https://llp.berkeley.edu/micropheno. Supplementary information: Supplementary data are available at Bioinformatics online.

SUMMARY: Metagenomics revolutionized the field of microbial ecology, giving access to Gb-sized datasets of microbial communities under natural conditions. This enables fine-grained analyses of the functions of community members, studies of their association with phenotypes and environments, as well as of their microevolution and adaptation to changing environmental conditions. However, phylogenetic methods for studying adaptation and evolutionary dynamics are not able to cope with big data. EDEN is the first software for the rapid detection of protein families and regions under positive selection, as well as their associated biological processes, from meta- and pangenome data. It provides an interactive result visualization for detailed comparative analyses. AVAILABILITY AND IMPLEMENTATION: EDEN is available as a Docker installation under the GPL 3.0 license, allowing its use on common operating systems, at http://www.github.com/hzi-bifo/eden. CONTACT: [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Characterizing the errors generated by common high-throughput sequencing platforms and telling true genetic variation from technical artefacts are two interdependent steps, essential to many analyses such as single nucleotide variant calling, haplotype inference, sequence assembly and evolutionary studies. Both random and systematic errors can show a specific occurrence profile for each of the six prominent sequencing platforms surveyed here: 454 pyrosequencing, Complete Genomics DNA nanoball sequencing, Illumina sequencing by synthesis, Ion Torrent semiconductor sequencing, Pacific Biosciences single-molecule real-time sequencing and Oxford Nanopore sequencing. There is a large variety of programs available for error removal in sequencing read data, which differ in the error models and statistical techniques they use, the features of the data they analyse, the parameters they determine from them and the data structures and algorithms they use. We highlight the assumptions they make and for which data types these hold, providing guidance which tools to consider for benchmarking with regard to the data properties. While no benchmarking results are included here, such specific benchmarks would greatly inform tool choices and future software development. The development of stand-alone error correctors, as well as single nucleotide variant and haplotype callers, could also benefit from using more of the knowledge about error profiles and from (re)combining ideas from the existing approaches presented here.

MOTIVATION: Gene assembly is an important step in functional analysis of shotgun metagenomic data. Nonetheless, strain aware assembly remains a challenging task, as current assembly tools often fail to distinguish among strain variants or require closely related reference genomes of the studied species to be available. RESULTS: We have developed Snowball, a novel strain aware gene assembler for shotgun metagenomic data that does not require closely related reference genomes to be available. It uses profile hidden Markov models (HMMs) of gene domains of interest to guide the assembly. Our assembler performs gene assembly of individual gene domains based on read overlaps and error correction using read quality scores at the same time, which results in very low per-base error rates. AVAILABILITY AND IMPLEMENTATION: The software runs on a user-defined number of processor cores in parallel, runs on a standard laptop and is available under the GPL 3.0 license for installation under Linux or OS X at https://github.com/hzi-bifo/snowball CONTACT: [email protected],[email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Metagenomics characterizes microbial communities by random shotgun sequencing of DNA isolated directly from an environment of interest. An essential step in computational metagenome analysis is taxonomic sequence assignment, which allows identifying the sequenced community members and reconstructing taxonomic bins with sequence data for the individual taxa. For the massive datasets generated by next-generation sequencing technologies, this cannot be performed with de-novo phylogenetic inference methods. We describe an algorithm and the accompanying software, taxator-tk, which performs taxonomic sequence assignment by fast approximate determination of evolutionary neighbors from sequence similarities. RESULTS: Taxator-tk was precise in its taxonomic assignment across all ranks and taxa for a range of evolutionary distances and for short as well as for long sequences. In addition to the taxonomic binning of metagenomes, it is well suited for profiling microbial communities from metagenome samples because it identifies bacterial, archaeal and eukaryotic community members without being affected by varying primer binding strengths, as in marker gene amplification, or copy number variations of marker genes across different taxa. Taxator-tk has an efficient, parallelized implementation that allows the assignment of 6 Gb of sequence data per day on a standard multiprocessor system with 10 CPU cores and microbial RefSeq as the genomic reference data.

2007.At a meeting at Rutgers University that summer, I convinced a very good postdoc, Ben Adams, who was working in Japan by that time, that joining my new lab would be a great idea.We arrived in Saarbrücken at about the same time in the fall, and that's how the

MOTIVATION: Ancestral character state reconstruction describes a set of techniques for estimating phenotypic or genetic features of species or related individuals that are the predecessors of those present today. Such reconstructions can reach into the distant past and can provide insights into the history of a population or a set of species when fossil data are not available, or they can be used to test evolutionary hypotheses, e.g. on the co-evolution of traits. Typical methods for ancestral character state reconstruction of continuous characters consider the phylogeny of the underlying data and estimate the ancestral process along the branches of the tree. They usually assume a Brownian motion model of character evolution or extensions thereof, requiring specific assumptions on the rate of phenotypic evolution. RESULTS: We suggest using ridge regression to infer rates for each branch of the tree and the ancestral values at each inner node. We performed extensive simulations to evaluate the performance of this method and have shown that the accuracy of its reconstructed ancestral values is competitive to reconstructions using other state-of-the-art software. Using a hierarchical clustering of gene mutation profiles from an ovarian cancer dataset, we demonstrate the use of the method as a feature selection tool. AVAILABILITY AND IMPLEMENTATION: The algorithm described here is implemented in C++ as a stand-alone program, and the source code is freely available at http://algbio.cs.uni-duesseldorf.de/software/RidgeRace.tar.gz. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Metagenome research uses random shotgun sequencing of microbial community DNA to study the genetic sequences of its members without cultivation. This development has been strongly supported by improvements in sequencing technologies, which have rendered sequencing cheaper than before. As a consequence, downstream computational analysis of metagenome sequence samples is now faced with large amounts of complex data. One of the essential steps in metagenome analysis is reconstruction of draft genomes for populations of a community or of draft 'pan-genomes' for higher level clades. 'Taxonomic binning' corresponds to the process of assigning a taxonomic identifier to sequence fragments, based on information such as sequence similarity, sequence composition or read coverage. This is used for draft genome reconstruction, if sequencing coverage is insufficient for reconstruction based on assembly information alone. Subsequent functional and metabolic annotation of draft genomes allows a genome-level analysis of novel uncultured microbial species and even inference of their cultivation requirements.

Distinguishing mutations that determine an organism's phenotype from (near-) neutral 'hitchhikers' is a fundamental challenge in genome research, and is relevant for numerous medical and biotechnological applications. For human influenza viruses, recognizing changes in the antigenic phenotype and a strains' capability to evade pre-existing host immunity is important for the production of efficient vaccines. We have developed a method for inferring 'antigenic trees' for the major viral surface protein hemagglutinin. In the antigenic tree, antigenic weights are assigned to all tree branches, which allows us to resolve the antigenic impact of the associated amino acid changes. Our technique predicted antigenic distances with comparable accuracy to antigenic cartography. Additionally, it identified both known and novel sites, and amino acid changes with antigenic impact in the evolution of influenza A (H3N2) viruses from 1968 to 2003. The technique can also be applied for inference of 'phenotype trees' and genotype-phenotype relationships from other types of pairwise phenotype distances.

BACKGROUND: Genome and metagenome studies have identified thousands of protein families whose functions are poorly understood and for which techniques for functional characterization provide only partial information. For such proteins, the genome context can give further information about their functional context. RESULTS: We describe a Bayesian method, based on a probabilistic topic model, which directly identifies functional modules of protein families. The method explores the co-occurrence patterns of protein families across a collection of sequence samples to infer a probabilistic model of arbitrarily-sized functional modules. CONCLUSIONS: We show that our method identifies protein modules - some of which correspond to well-known biological processes - that are tightly interconnected with known functional interactions and are different from the interactions identified by pairwise co-occurrence. The modules are not specific to any given organism and may combine different realizations of a protein complex or pathway within different taxa.

MOTIVATION: As a starting point in annotation of bacterial genomes, gene finding programs are used for the prediction of functional elements in the DNA sequence. Due to the faster pace and increasing number of genome projects currently underway, it is becoming especially important to have performant methods for this task. RESULTS: This study describes the development of joint application strategies that combine the strengths of two microbial gene finders to improve the overall gene finding performance. Critica is very specific in the detection of similarity-supported genes as it uses a comparative sequence analysis-based approach. Glimmer employs a very sophisticated model of genomic sequence properties and is sensitive also in the detection of organism-specific genes. Based on a data set of 113 microbial genome sequences, we optimized a combined application approach using different parameters with relevance to the gene finding problem. This results in a significant improvement in specificity while there is similarity in sensitivity to Glimmer. The improvement is especially pronounced for GC rich genomes. The method is currently being applied for the annotation of several microbial genomes. AVAILABILITY: The methods described have been implemented within the gene prediction component of the GenDB genome annotation system.