Search

Robust annotation of metabolites is a challenging task in metabolomics. Among available applications, 13C NMR experiment INADEQUATE determines direct 13C-13C connectivity unambiguously, offering indispensable information on molecular structure. Despite its great utility, it is not always practical to collect INADEQUATE data on every sample in a large metabolomics study because of its relatively long experiment time. Here, we propose an alternative approach that maintains the quality of information but saves experiment time. In this approach, individual samples in a study are first screened by 13C homonuclear J-resolved experiment (JRES). Next, JRES data are processed by statistical total correlation spectroscopy (STOCSY) to extract peaks that behave similarly among samples. Finally, INADEQUATE is collected on one internal pooled sample to select STOCSY peaks that originate from the same compound. We tested this concept using the 13C-labeled endometabolome of a model marine diatom strain incu-bated under various settings, intending to cover a range of metabolites produced under different external conditions. This scheme was able to extract known diatom metabolites proline, 2,3-dihydroxypropane-1-sulfonate (DHPS), b-1,3-glucan, choline, and glutamate. This pipeline also detected unknown compounds with structural information, which is valuable in metabolomics where a priori knowledge of metabolites is not always available. The ability of this scheme was seen even in sugar regions, which are usually chal-lenging in 1H NMR due to severe peak overlap. JRES and INADEQUATE were highly complementary; INADEQUATE provided directly-bonded 13C networks, whereas JRES linked INADEQUATE networks within the same compound but broken by nitrogen or sulfur atoms, highlighting the advantage of this integrated approach.(c) 2023 The Authors. Published by Elsevier Inc. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Recent developments in deep learning, coupled with an increasing number of sequenced proteins, have led to a breakthrough in life science applications, in particular in protein property prediction. There is hope that deep learning can close the gap between the number of sequenced proteins and proteins with known properties based on lab experiments. Language models from the field of natural language processing have gained popularity for protein property predictions and have led to a new computational revolution in biology, where old prediction results are being improved regularly. Such models can learn useful multipurpose representations of proteins from large open repositories of protein sequences and can be used, for instance, to predict protein properties. The field of natural language processing is growing quickly because of developments in a class of models based on a particular model—the Transformer model. We review recent developments and the use of large-scale Transformer models in applications for predicting protein characteristics and how such models can be used to predict, for example, post-translational modifications. We review shortcomings of other deep learning models and explain how the Transformer models have quickly proven to be a very promising way to unravel information hidden in the sequences of amino acids.

Recent years have witnessed the detection of an increasing number of complex organic molecules in interstellar space, some of them being of prebiotic interest. Disentangling the origin of interstellar prebiotic chemistry and its connection to biochemistry and ultimately to biology is an enormously challenging scientific goal where the application of complexity theory and network science has not been fully exploited. Encouraged by this idea, we present a theoretical and computational framework to model the evolution of simple networked structures toward complexity. In our environment, complex networks represent simplified chemical compounds, and interact optimizing the dynamical importance of their nodes. We describe the emergence of a transition from simple networks toward complexity when the parameter representing the environment reaches a critical value. Notably, although our system does not attempt to model the rules of real chemistry, nor is dependent on external input data, the results describe the emergence of complexity in the evolution of chemical diversity in the interstellar medium. Furthermore, they reveal an as yet unknown relationship between the abundances of molecules in dark clouds and the potential number of chemical reactions that yield them as products, supporting the ability of the conceptual framework presented here to shed light on real scenarios. Our work reinforces the notion that some of the properties that condition the extremely complex journey from the chemistry in space to prebiotic chemistry and finally to life could show relatively simple and universal patterns., Comment: Published in PNAS. The Supplementary Information can be found at https://github.com/MiguelGarciaSanchez/NetWorld/blob/main/garcia_sanchez_et_al_Supp_Info_2022.pdf

Recent advances in sequencing and bioinformatics have expanded the tree of life by providing genomes for uncultured environmentally relevant clades, either through metagenome-assembled genomes or through single-cell genomes. While this expanded diversity can provide novel insights into microbial population structure, most tools available for core-genome estimation are sensitive to genome completeness. Consequently, a major portion of the huge phylogenetic diversity uncovered by environmental genomic approaches remains excluded from such analyses. We present mOTUpan, a novel iterative Bayesian method for computing the core genome for sets of genomes of highly diverse completeness range. The likelihood for each gene cluster to belong to core or accessory genome is estimated by computing the probability of its presence/absence pattern in the target genome set. The core-genome prediction is computationally efficient and can be scaled up to thousands of genomes. It has shown comparable estimates to state-of-the-art tools Roary and PPanGGOLiN for high-quality genomes and is capable of using genomes at lower completeness thresholds. mOTUpan wraps a bootstrapping procedure to estimate the quality of a specific core-genome prediction, as the accuracy of each run will depend on the specific completeness distribution and the number of genomes in the dataset under scrutiny. mOTUpan is implemented in the mOTUlizer software package, and available at github.com/moritzbuck/mOTUlizer, under GPL 3.0 license.

Multi-trait and multi-environment analyses can improve genomic prediction by exploiting between-trait correlations and genotype-by-environment interactions. In the context of reaction norm models, genotype-by-environment interactions can be described as functions of high-dimensional sets of markers and environmental covariates. However, comprehensive multi-trait reaction norm models accounting for marker × environmental covariates interactions are lacking. In this article, we propose to extend a reaction norm model incorporating genotype-by-environment interactions through (co)variance structures of markers and environmental covariates to a multi-trait reaction norm case. To do that, we propose a novel methodology for characterizing the environment at different growth stages based on growth degree-days (GDD). The proposed models were evaluated by variance components estimation and predictive performance for winter wheat grain yield and protein content in a set of 2,015 F6-lines. Cross-validation analyses were performed using leave-one-year-location-out (CV1) and leave-one-breeding-cycle-out (CV2) strategies. The modeling of genomic [SNPs] × environmental covariates interactions significantly improved predictive ability and reduced the variance inflation of predicted genetic values for grain yield and protein content in both cross-validation schemes. Trait-assisted genomic prediction was carried out for multi-trait models, and it significantly enhanced predictive ability and reduced variance inflation in all scenarios. The genotype by environment interaction modeling via genomic [SNPs] × environmental covariates interactions, combined with trait-assisted genomic prediction, boosted the benefits in predictive performance. The proposed multi-trait reaction norm methodology is a comprehensive approach that allows capitalizing on the benefits of multi-trait models accounting for between-trait correlations and reaction norm models exploiting high-dimensional genomic and environmental information.

AcetoBase is a public repository and database of formyltetrahydrofolate synthetase (FTHFS) sequences. It is the first systematic collection of bacterial FTHFS nucleotide and protein sequences from genomes and metagenome-assembled genomes and of sequences generated by clone library sequencing. At its publication in 2019, AcetoBase (Version 1) was also the first database to establish connections between the FTHFS gene, the Wood–Ljungdahl pathway and 16S ribosomal RNA genes. Since the publication of AcetoBase, there have been significant improvements in the taxonomy of many bacterial lineages and accessibility/availability of public genomics and metagenomics data. The update to the AcetoBase reference database described here (Version 2) provides new sequence data and taxonomy, along with improvements in web functionality and user interface. The evaluation of this latest update by re-analysis of publicly accessible FTHFS amplicon sequencing data previously analysed with AcetoBase Version 1 revealed significant improvements in the taxonomic assignment of FTHFS sequences. Database URL: https://acetobase.molbio.slu.se

For multivariate data analysis involving only two input matrices, the previously published methods for variable influence on projection (e.g., VIPOPLS or VIPO2PLS) are widely used for variable selection purposes, including (i) variable importance assessment, (ii) dimensionality reduction of big data and (iii) interpretation enhancement of PLS, OPLS and O2PLS models. For multiblock analysis, the OnPLS models find relationships among multiple data matrices by calculating latent variables; however, a method for improving the interpretation of these latent variables by assessing the importance of the input variables was not available up to now. A method for variable selection in multiblock analysis, called multiblock variable influence on orthogonal projections (MB-VIOP) is explained in this paper. MB-VIOP is a model based variable selection method that uses the data matrices, the scores and the normalized loadings of an OnPLS model in order to sort the input variables of more than two data matrices according to their importance for both simplification and interpretation of the total multiblock model, and also of the unique, local and global model components separately. MB-VIOP has been tested using three multiblock datasets. MB-VIOP assesses the variable importance in any type of data. MB-VIOP connects the input variables of different data matrices according to their relevance for the interpretation of each latent variable, yielding enhanced interpretability for each OnPLS model component. Besides, MB-VIOP can deal with strong overlapping of types of variation, as well as with many data blocks with very different dimensionality. The ability of MB-VIOP for generating dimensionality reduced models with high interpretability makes this method ideal for big data mining, multi-omics data integration and any study that requires exploration and interpretation of large streams of data., It has 54 pages (including 4 figures, 5 tables, and supporting information). Submitted to peer-reviewed journal

Degradome sequencing is commonly used to generate high-throughput information on mRNA cleavage sites mediated by small RNAs (sRNA). In our datasets of potato (Solanum tuberosum, St) and Phytophthora infestans (Pi), initial predictions generated high numbers of cleavage site predictions, which highlighted the need of improved analytic tools. Here, we present an R package based on a deep learning convolutional neural network (CNN) in a machine learning environment to optimize discrimination of false from true cleavage sites. When applying smartPARE to our datasets on potato during the infection process by the late blight pathogen, 7.3% of all cleavage windows represented true cleavages distributed on 214 sites in P. infestans and 444 sites in potato. The sRNA landscape of the two organisms is complex with uneven sRNA production and cleavage regions widespread in the two genomes. Multiple targets and several cases of complex regulatory cascades, particularly in potato, was revealed. We conclude that our new analytic approach is useful for anyone working on complex biological systems and with the interest of identifying cleavage sites particularly inferred by sRNA classes beyond miRNAs.

Background Translation is a tightly regulated process, controlling the rate of protein synthesis in cells. Ribosome sequencing (Ribo-Seq) is a recently developed tool for studying actively translated mRNA and can thus directly address translational regulation. Ribo-Seq libraries need to be sequenced to a great depth due to high contamination by rRNA and other contaminating nucleic acid fragments. Deep sequencing is expensive, and it generates large volumes of data, making data analysis complicated and time consuming. Methods and results Here we developed a platform for Ribo-Seq library construction and data analysis to enable rapid quality assessment of Ribo-Seq libraries with the help of a small-scale sequencer. Our data show that several qualitative features of a Ribo-Seq library, such as read length distribution, P-site distribution, reading frame and triplet periodicity, can be effectively evaluated using only the data generated by a benchtop sequencer with a very limited number of reads. Conclusion Our pipeline enables rapid evaluation of Ribo-Seq libraries, opening up possibilities for optimization of Ribo-Seq library construction from difficult samples, and leading to better decision making prior to more costly deep sequencing.

Metagenomics has redefined many areas of microbiology. However, metagenome-assembled genomes (MAGs) are often fragmented, primarily when sequencing was performed with short reads. Recent long-read sequencing technologies promise to improve genome reconstruction. However, the integration of two different sequencing modalities makes downstream analyses complex. We, therefore, developed MUFFIN, a complete metagenomic workflow that uses short and long reads to produce high-quality bins and their annotations. The workflow is written by using Nextflow, a workflow orchestration software, to achieve high reproducibility and fast and straightforward use. This workflow also produces the taxonomic classification and KEGG pathways of the bins and can be further used for quantification and annotation by providing RNA-Seq data (optionally). We tested the workflow using twenty biogas reactor samples and assessed the capacity of MUFFIN to process and output relevant files needed to analyze the microbial community and their function. MUFFIN produces functional pathway predictions and, if provided de novo metatranscript annotations across the metagenomic sample and for each bin. MUFFIN is available on github under GNUv3 licence: https://github.com/RVanDamme/MUFFIN., Author summary Determining the entire DNA of environmental samples (sequencing) is a fundamental approach to gain deep insights into complex bacterial communities and their functions. However, this approach produces enormous amounts of data, which makes analysis time intense and complicated. We developed the Software “MUFFIN,” which effortlessly untangle the complex sequencing data to reconstruct individual bacterial species and determine their functions. Our software is performing multiple complicated steps in parallel, automatically allowing everyone with only basic informatics skills to analyze complex microbial communities. For this, we combine two sequencing technologies: "long-sequences" (nanopore, better reconstruction) and "short-sequences" (Illumina, higher accuracy). After the reconstruction, we group the fragments that belong together ("binning") via multiple approaches and refinement steps while also utilizing the information from other bacterial communities ("differential binning"). This process creates hundreds of "bins" whereas each represents a different bacterial species with a unique function. We automatically determine their species, assess each genome’s completeness, and attribute their biological functions and activity ("transcriptomics and pathways"). Our Software is entirely freely available to everyone and runs on a good computer, compute cluster, or via cloud.

We develop a Bayesian model (BayesRR-RC) that provides robust SNP-heritability estimation, an alternative to marker discovery, and accurate genomic prediction, taking 22 seconds per iteration to estimate 8.4 million SNP-effects and 78 SNP-heritability parameters in the UK Biobank. We find that only ≤10% of the genetic variation captured for height, body mass index, cardiovascular disease, and type 2 diabetes is attributable to proximal regulatory regions within 10kb upstream of genes, while 12-25% is attributed to coding regions, 32–44% to introns, and 22-28% to distal 10-500kb upstream regions. Up to 24% of all cis and coding regions of each chromosome are associated with each trait, with over 3,100 independent exonic and intronic regions and over 5,400 independent regulatory regions having ≥95% probability of contributing ≥0.001% to the genetic variance of these four traits. Our open-source software (GMRM) provides a scalable alternative to current approaches for biobank data., Improving inference in large-scale genetic data linked to electronic medical record data requires the development of novel computationally efficient regression methods. Here, the authors develop a Bayesian approach for association analyses to improve SNP-heritability estimation, discovery, fine-mapping and genomic prediction.

Background: Anaerobic digestion (AD) has long been critical technology for green energy, but the majority of the microorganisms involved are unknown and are currently not cultivable, which makes abundance tracking difficult. Developments in nanopore long-read sequencing make it a promising approach for monitoring microbial communities via metagenomic sequencing. For reliable monitoring of AD via long reads, a robust protocol for obtaining less fragmented, high-quality DNA, while preserving bacteria and archaea composition, was established. Results: Samples from 20 different biogas/wastewater reactors were investigated, and a median of 20.5 Gb sequencing data per nanopore flow cell was retrieved for each reactor using the developed DNA isolation protocol. The nanopore sequencing data was compared against Illumina sequencing data while using different taxonomic indices for read classifications. The Genome Taxonomy Database (GTDB) index allowed sufficient characterisation of the abundance of bacteria and archaea in biogas reactors with a dramatic improvement (1.8- to 13-fold increase) in taxonomic classification compared to the RefSeq index. Both technologies performed similarly in taxonomic read classification with a slight advantage for Illumina in regards to the total proportion of classified reads. However, nanopore sequencing data revealed a higher genus richness after classification. Conclusion: Metagenomic read classification via nanopore provides a promising approach to monitor the abundance of taxa present in a microbial AD community, as an alternative to 16S rRNA studies or Illumina Sequencing.

Inherited retinal degenerations (IRDs) form a clinically and genetically heterogeneous group of diseases, leading to visual impairment or blindness in both humans and dogs. The prevalence of IRDs is estimated at 1 in 2,000 in humans. In dogs, the exact prevalence is unknown, but close to 100 different breeds have been reported to be affected, many by more than one type of IRD. The identification of the underlying genetic variants is critical, as the results can be used to develop genetic tests, which allow breeders to make informed breeding decisions while preserving genetic variation. In Labrador retrievers, a novel form of IRD was recently identified, with clinical signs indicating cone-rod photoreceptor degeneration. In this thesis, a whole-genome sequencing approach was used to identify a frameshift insertion leading to a premature stop codon in the canine ABCA4 gene. In humans, mutations in the ABCA4 gene are the major cause of Stargardt disease (STGD), an autosomal recessive retinal degeneration leading to central visual impairment. Transcript and protein level investigations showed that the canine ABCA4 insertion is a loss-of-function mutation responsible for the novel canine IRD, and leads to a phenotype similar to STGD in humans. Golden retrievers are affected by at least four different forms of IRD, one of which is associated with a deletion in the TTC8 gene. Mutations in this gene in humans are involved in the Bardet-Biedl syndrome (BBS) with heterogeneous clinical signs. We were able to show that the canine deletion is a loss-of-function mutation resulting in a syndromic IRD similar to BBS. Lastly, while the human retinal transcriptome has been extensively studied, less is known about the gene expression patterns in the canine retina. Using short- and longread cDNA sequencing we characterized the canine retinal transcriptome, results that in the future can be used to identify and validate causative genetic variants for canine IRDs. The results of this thesis contribute to the understanding of two important IRDs affecting the health and welfare of both dogs and humans. In addition, the thesis highlights the importance of a well-characterized retinal transcriptome for successful identification of disease-causing alleles.

This deliverable report describes the applicability of the Minimalist Mixture Model – M3 – as a prototype tool for predicting performance of innovative plant teams. M3 has been developed as part of the work conducted in DIVERSify within WP3. M3, and its most recent further developments, are briefly described. The emphasis is on the general structure of the model, and the data needed for its calibration, running and validation. In addition, this report presents two applications of M3 to predict the performance of existing and novel plant teams. In the first application, M3 is used to determine the performance of plant teams differing from the existing ones for specific parameters, thus helping in the identification of key traits for superior performance, and hence potential innovative plant teams that should be prioritized in breeding. In the second application, M3 is used to determine the stability of a plant team performance in the face of variable climatic conditions, including those projected for the future.

Genome-wide prediction (GWP) has become the state-of-the art method in artificial selection. Data sets often comprise number of genomic markers and individuals in ranges from a few thousands to millions. Hence, computational efficiency is important and various machine learning methods have successfully been used in GWP. Neural networks (NN) and deep learning (DL) are very flexible methods that usually show outstanding prediction properties on complex structured data, but their use in GWP is nevertheless rare and debated. This study describes a powerful NN method for genomic marker data that can easily be extended. It is shown that a one-dimensional convolutional neural network (CNN) can be used to incorporate the ordinal information between markers and, together with pooling and l (1)-norm regularization, provides a sparse and computationally efficient approach for GWP. The method, denoted CNNGWP, is implemented in the deep learning software Keras, and hyper-parameters of the NN are tuned with Bayesian optimization. Model averaged ensemble predictions further reduce prediction error. Evaluations show that CNNGWP improves prediction error by more than 25% on simulated data and around 3% on real pig data compared with results obtained with GBLUP and the LASSO. In conclusion, the CNNGWP provides a promising approach for GWP, but the magnitude of improvement depends on the genetic architecture and the heritability.

Examination of closely related species pairs is suggested for evolutionary comparisons of different degrees of polyphagy, which we did here with three taxa of lepidopteran herbivores, Spodoptera spp (S. littoralis, S. frugiperda maize (C) and rice (R) strains) for a RNAseq analysis of the midguts from the 3rd instar insect larvae for differential metabolic responses after feeding on pinto bean based artificial diet vs maize leaves. Paired-end (2×100 bp) Illumina HiSeq2500 sequencing resulted in a total of 24, 23, 24, and 21 million reads for the SF-C-Maize, SF-C-Pinto, SF-R-Maize, SF-R Pinto, and a total of 35 and 36 million reads for the SL-Maize and SL-Pinto samples, respectively. After quality control measures, a total of 62.2 million reads from SL and 71.7 million reads from SF were used for transcriptome assembly (TA). The resulting final de novo reference TA (backbone) for the SF taxa contained 37,985 contigs with a N50 contig size of 1030 bp and a maximum contig length of 17,093 bp, while for SL, 28,329 contigs were generated with a N50 contig size of 1980 bp and a maximum contig length of 18,267 bp. The data presented herein contains supporting information related to our research article Roy et al. (2016) http://dx.doi.org/10.1016/j.ibmb.2016.02.006[1]., Graphical abstract fx1

Type 2 diabetes (T2D) is a major global health problem and prevention could be improved by identifying individuals at risk at an early stage, followed by preventive strategies, e.g., dietary modifications. Untargeted LC-MS metabolomics offers the possibility to identify predictive biomarkers that may improve risk prediction and dietary biomarkers that may facilitate investigation of diet-T2D relationships. However, untargeted metabolomics generates large-scale data, resulting in demanding data processing and statistical analyses preceding meaningful biological interpretation. The work presented in this thesis sought to develop bioinformatics tools for dealing with large-scale data generated from untargeted LC-MS metabolomics, to apply such tools to identify predictive metabolites of T2D and metabolites related to predefined healthy Nordic dietary indices, and to investigate whether such metabolites are associated with T2D risk in a Swedish population. Two novel R programming based packages were developed: ‘batchCorr’, a data-processing strategy to correct for within- and between-batch variability in LC-MS experiments, and ‘MUVR’, a statistical framework for multivariate analysis with unbiased variable selection. These tools were applied on untargeted LC-MS metabolomics data obtained from plasma samples from a nested case-control study. Overall, 46 predictive metabolites of T2D were identified. Several metabolites showed good long-term reproducibility among healthy participants, reinforcing their potential as predictive biomarkers, while some changed in the disease-associated direction among cases, reflecting disease progression. In total, 38 metabolites were found to be associated with two predefined healthy Nordic dietary indices. No evidence was found to support association between indices and T2D risk. Instead, metabolites related to unhealthy foods not captured in indices were associated with increased risk. In conclusion, the novel bioinformatics tools developed here can overcome vital data-analytical challenges inherent in large-scale untargeted metabolomics studies. Predictive metabolites have great potential to provide information related to T2D pathophysiology and monitoring of disease progression, though only a limited improvement in disease prediction was achieved when adding them to models based on optimally selected traditional risk factors. Moreover, no evidence was found of an association between healthy Nordic dietary indices and T2D risk. Future studies should investigate how diet/lifestyle risk factors affect pathological pathways of T2D and prevent disease development by integration of multi-omics techniques and traditional methods.

Modern technology allows researchers to generate data at an ever increasing rate, outpacing the capacity of researchers to analyse it. Developing automated support systems for the collection, management and distribution of information is therefore an important step to reduce error rates and accelerate progress to enable high-quality research based on big data volumes. This thesis encompasses five articles, describing strategies for the creation of technical research platforms, as well as descriptions of the technical platforms themselves. The key conclusion of the thesis is that technical solutions for many issues have been available for a long time. These technical solutions are however overlooked, or simply ignored, if they fail to recognise the social dimensions of the issues they try to solve. The Molecular Methods database is an example of a technically sound but only partially successful solution in regards to social viability. Thousands of researchers have used the website to access protocols, but only a handful have shared their own work on MolMeth. Experiences from the Molecular Methods database and other projects have provided a foundation for studies supporting the development of the eB3Kit The eB3Kit is a portable, robust and scalable informatics platform for structured data management. Deploying the platform enables research groups to carry out advanced research projects with very limited means. With the eB3Kit researchers can integrate data from a wide variety of sources, including the local laboratory information management system and analyse it using the Galaksio interface. Galaksio provides user friendly access to the Galaxy workflow management system and provides eB3Kit users with access to tools developed by a far larger user community than the one actively developing the eB3Kit. Using a workflow management system improves reproducibility and enables bioinformaticians to prepare workflows without directly accessing ethically or commercially sensitive data. Therefore, it is especially well- suited for applications where researchers are worried about privacy and during disease outbreaks where persistent storage and analysis capacity must be established quickly.

Massive parallel sequencing technologies have made a remarkable contribution in understanding the genomic complexity in human and other mammalian species. This thesis includes four studies illustrating the technology change from short-read to long-read massive parallel sequencing and the corresponding layers of genomic complexity resolved in dog and horse. In the first study, the Ion Proton sequencing platform was evaluated for whole-genome re-sequencing of four dog genomes. On average, 80 % of the genome and 77 % of the exome had at least 4-fold coverage. The obtained genotypes showed 90 % concordance with those using the CanineHD BeadChip. Next, whole-genome re-sequencing of a canine family trio was used as a proof-of-concept to map an autosomal recessive disorder termed hereditary footpad hyperkeratosis in the Kromfohrländer breed. This resulted in the identification and association of a single nucleotide variant in the FAM83G gene. In the third study, we generated the first long-read assembly of the horse MHC class II region (1.2 Mb) by assembling eight bacterial artificial chromosome clones sequenced with single molecule, real-time (SMRT) sequencing technology from Pacific Biosciences (PacBio). Detailed annotation of this assembly revealed 23 functional genes and 12 pseudogenes. In comparison to other mammals, the horse MHC showed an increased number of MHC class II genes in different relative positions and with different directionality. In the fourth study, “sequence self-similarity” of the equine MHC class II region was investigated to identify paralogous segments and to reconstruct the evolutionary history of duplication events. The results indicated a major expansion of class II loci 55-70 million years ago, coinciding with mammalian radiation. We showed that the high genomic structural diversity and plasticity of the mammalian MHC region may be a result of recurring segmental duplication events mediated either by unequal crossing-over between misaligned genomic regions or, possibly, by alternative transposition of long interspersed nuclear elements.

Through the influence of abiotic factors, the habitat use of organisms affects their metabolism as well as other species- and size-dependent individual-based rates. The habitat-specific performances of individuals interacting in different habitats thereby affect biotic interactions. Habitat use is thus central for the outcomes of biotic interactions that, in turn, regulate populations and communities. My aim is to investigate how individual processes are influenced by habitat-dependent abiotic factors, affecting biotic interactions to regulate habitat use and population structures in fish communities. I examined patterns of habitat distribution and population structures of perch (Perca fluviatilis L.), roach (Rutilus rutilus (L.)), and the zooplankton specialist vendace (Coregonus albula (L.)) using a database of standardised test fishing data in lakes. To clarify mechanisms, I experimentally studied predation from perch in pond enclosures as well as relative foraging abilities of the two competitors roach and vendace in aquaria with different temperature and light treatments. To test mechanisms in natural situations, I calculated species- and size-dependent net energy intake, incorporating temperature- and light-dependence, including metabolism, using field data from different habitats in lakes with and without vendace. I also developed and applied a stage-structured biomass model, considering a cold water species (vendace) using two habitats differing in temperature. I thereby studied how climate warming which acts differently on different lake habitats affected temperature-dependent individual-based processes, and results on the population level. Through multi-species studies, I found that a combination of size- and environment-dependent individual processes determining energy gain, rather than predation risk, could explain size- and species-specific habitat use. The single-species study showed that stage-specific intake rates in one habitat, altered by increased temperature, affected intraspecific competition in both habitats, through a mechanism of ‘inter-habitat subsidies’ which altered population structure through maturation and reproduction rates. My thesis shows how including size- and environment-dependent individual processes, and interactions across habitats, increases our understanding of population and community structure as well as effects of environmental change.

Metagenomics, the sequence characterization of all genomes within a sample, is widely used as a virus discovery tool as well as a tool to study viral diversity of animals. Metagenomics can be considered to have three main steps; sample collection and preparation, sequencing and finally bioinformatics. Bioinformatic analysis of metagenomic datasets is in itself a complex process, involving few standardized methodologies, thereby hampering comparison of metagenomics studies between research groups. In this publication the new bioinformatics framework MetLab is presented, aimed at providing scientists with an integrated tool for experimental design and analysis of viral metagenomes. MetLab provides support in designing the metagenomics experiment by estimating the sequencing depth needed for the complete coverage of a species. This is achieved by applying a methodology to calculate the probability of coverage using an adaptation of Stevens’ theorem. It also provides scientists with several pipelines aimed at simplifying the analysis of viral metagenomes, including; quality control, assembly and taxonomic binning. We also implement a tool for simulating metagenomics datasets from several sequencing platforms. The overall aim is to provide virologists with an easy to use tool for designing, simulating and analyzing viral metagenomes. The results presented here include a benchmark towards other existing software, with emphasis on detection of viruses as well as speed of applications. This is packaged, as comprehensive software, readily available for Linux and OSX users at https://github.com/norling/metlab.

The camel is the most valuable livestock species in arid and semi-arid regions in the Greater Horn of Africa. Streptococcus agalactiae and Staphylococcus aureus are important pathogens for a wide range of hosts including camels, cattle and humans. Streptococcus agalactiae has been reported to cause infections of the skin, the respiratory tract, the mammary gland and the vaginal tract in camels. Staphylococcus aureus has been isolated from the nasal cavity, wound infections and mastitis from camels. Both pathogens account for decline in health and productivity of camels, hence causing economic losses to the inhabitants of arid and semi arid lands. To define candidate virulence traits in these bacteria, we compared the genomes of S. agalactiae and S. aureus. We sequenced and completely assembled the genomes of two S. agalactiae isolates ILRI005 and ILRI112 from abscessed case camels and an S. aureus isolate ILRI_Eymole1/1 from the nasal swab of camel in Kenya. To perform comparative analysis, we also sequenced and assembled an S. agalactiae isolate 09mas018883 from subclinical mastitis case cattle in Sweden. Mapping assembly, de novo assembly and post-assembly genome finishing were performed to obtain completely assembled genomes. Comparative genomics approach was applied to explore the genetic heterogeneity, core genome construction and protein repertoire comparison of these novel genomes, and to highlight potential virulence factors that could have contributed to the pathogenicity of these isolates in their hosts. Newly sequenced camel S. agalactiae genomes were compared with human and cattle S. agalactiae genomes. This comparison revealed that the two camel isolates were genetically close to each other but relatively distinct from other isolates, while cattle isolate 09mas018883 was genetically closer to the human isolates. Large proportion of the isolate-specific genes of the camel S. agalactiae isolates was clustered in putative phage insertions and genomic islands suggesting the lateral transfer of these putative phages. The two camel S. agalactiae isolates shared a novel potential virulent locus, the CRISPR2 (Cluster Regularly Interspaced Palindromic Repeats) locus. The two cattle S. agalactiae isolates and three human S. agalactiae isolates contained similar putative phage insertions. Important potential pathogenic factors found in all S. agalactiae isolates were CRISPR1 locus, cyl locus, capsular polysaccharide locus and pilus islands. Phylogenetic analysis of novel camel S. aureus genome of strain type ST30 and previously sequenced human S. aureus genomes of type Clonal Complex 30 (CC30) revealed that camel S. aureus isolate is genetically distinct from human S. aureus isolates of the same sequence type. Important features were also identified such as genes encoding bacterial adhesins and secretory proteins. The availability of genomic sequences of S. agalactiae and S. aureus from camels, their detailed bioinformatics analysis and identified potential virulence factors will foster the development of control measures such as molecular diagnostic assays and vaccines for control of S. agalactiae and S. aureus infections in camels. This will ensure improvement in health and productivity of camels.

Spirochaetal bacteria are highly diverse in terms of life style, growth requirements and virulence, but they all belong to a monophyletic ancient lineage. Many spirochaetes are difficult to culture or have fastidious growth requirements. Genus Brachyspira and genus Treponema, are two important members of phylum spirochetes that include well known pathogens, such as T.pallidum subsp. pallidum and B. hyodysenteriae. Other members of the phylum include T.pedis, associated with both bovine digital dermatitis (BDD) in cattle and necrotic skin lesions in pigs, T.phagedenis associated with BDD, and "B.suanatina", which has been reported in connection with swine dysentery-like enteric disease in pigs. All these bacteria are thus associated with significant health problems in farm animals leading to economic losses. The aims of this thesis were to identify and characterize potential pathogenic factors in T.pedis and T. phagedenis and to study the genomic characteristics of B. suanatina for a complete species validation and description. Assembly, annotation and bioinformatics analysis of the genome sequence of T. phagedenis strain V1 and T. pedis strain TA4 were performed. The complete genome of T. pedis strain TA4 was compared to that of T. denticola, a species associated with human periodontal disease. Results of the analysis showed close relatedness between these two bacterial species. Some of the pathogenicity related genes that are already known in T. denticola were also found in T. pedis strain TA4. In the genome of T. phagedenis strain V1, homologous proteins to known pathogenicity factors in T. denticola and T. pallidum were found and a locus encoding antigenic lipoproteins with potential for antigenic variation was characterized To identify the taxonomical position of B. suanatina, we performed assembly, annotation and phylogenetic analysis of the B. suanatina strain AN4859/03 genome along with its comparison with B. hyodysenteriae and other Brachyspira species. Comparative analysis suggested that B. suanatina is a novel species, though phenotypically it showed no difference from B. hyodysenteriae.

Nitrous oxide (N₂O) is a potent greenhouse gas and the major ozone depleting substance in the stratosphere. One major source of N₂O is incomplete denitrification, whereas the only known tropospheric sink of N₂O is the microbial enzyme nitrous oxide reductase. Denitrification is defined as the stepwise reduction of nitrite to dinitrogen via nitric oxide and N₂O by facultative anaerobic microorganisms. This thesis aims to elucidate the phylogenetic diversity of the N₂O reductase encoding gene nosZ and its context in microbial genomes in relation to other genes in the denitrification pathway, as well as the relative influence of plants and soil on the activity, abundance and structure of N₂O-reducing communities in the rhizosphere. Phylogenetic analysis of publicly available nosZ gene sequences revealed that its genetic diversity is divided into two distinct clades termed clade I and clade II, the latter having not been accounted for in previous studies. Newly developed molecular tools revealed that it is abundant in a wide range of environments. Analysis of microbial genomes showed that co-occurrence patterns of nosZ with other denitrification genes were neither randomly distributed among taxonomic units nor among habitats. Many genomes had truncated pathways as organisms possessing nosZII often lacked other genes involved in denitrification, suggesting these organisms may act as N₂O-sinks in the environment. Pot experiments with sunflower and barley indicated a niche differentiation between the two nosZ gene variants, as nosZI showed an affinity for plant roots while nosZII was more abundant in the surrounding soil. However, denitrification and N₂O-production activity in soil were controlled by edaphic factors. Moreover, an intercropping experiment with cocksfoot and lucerne showed that intercropping had a negative influence on nosZII abundances on cocksfoot roots which in conjunction with phylogenetic placement of sequencing reads indicated the presence of organisms with only nosZ lacking a denitrification pathway. In conclusion, the development of new molecular tools combined with comparative genomic analysis sheds new light on the ecology of biological N₂O reduction in the rhizosphere.

Background Amplicon sequencing is an established and cost-efficient method for profiling microbiomes. However, many available tools to process this data require both bioinformatics skills and high computational power to process big datasets. Furthermore, there are only few tools that allow for long read amplicon data analysis. To bridge this gap, we developed the LotuS2 (less OTU scripts 2) pipeline, enabling user-friendly, resource friendly, and versatile analysis of raw amplicon sequences. Results In LotuS2, six different sequence clustering algorithms as well as extensive pre- and post-processing options allow for flexible data analysis by both experts, where parameters can be fully adjusted, and novices, where defaults are provided for different scenarios. We benchmarked three independent gut and soil datasets, where LotuS2 was on average 29 times faster compared to other pipelines, yet could better reproduce the alpha- and beta-diversity of technical replicate samples. Further benchmarking a mock community with known taxon composition showed that, compared to the other pipelines, LotuS2 recovered a higher fraction of correctly identified taxa and a higher fraction of reads assigned to true taxa (48% and 57% at species; 83% and 98% at genus level, respectively). At ASV/OTU level, precision and F-score were highest for LotuS2, as was the fraction of correctly reported 16S sequences. Conclusion LotuS2 is a lightweight and user-friendly pipeline that is fast, precise, and streamlined, using extensive pre- and post-ASV/OTU clustering steps to further increase data quality. High data usage rates and reliability enable high-throughput microbiome analysis in minutes. Availability LotuS2 is available from GitHub, conda, or via a Galaxy web interface, documented at http://lotus2.earlham.ac.uk/.