MOTIVATION: Recovery of metagenome-assembled genomes (MAGs) from shotgun metagenomic data is an important task for the comprehensive analysis of microbial communities from variable sources. Single binning tools differ in their ability to leverage specific aspects in MAG reconstruction, the use of ensemble binning refinement tools is often time consuming and computational demand increases with community complexity. We introduce MAGScoT, a fast, lightweight and accurate implementation for the reconstruction of highest-quality MAGs from the output of multiple genome-binning tools. RESULTS: MAGScoT outperforms popular bin-refinement solutions in terms of quality and quantity of MAGs as well as computation time and resource consumption. AVAILABILITY AND IMPLEMENTATION: MAGScoT is available via GitHub (https://github.com/ikmb/MAGScoT) and as an easy-to-use Docker container (https://hub.docker.com/repository/docker/ikmb/magscot). SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Reference-based phasing and genotype imputation algorithms have been developed with sublinear theoretical runtime behaviour, but runtimes are still high in practice when large genome-wide reference datasets are used. RESULTS: We developed EagleImp, a software based on the methods used in the existing tools Eagle2 and PBWT, which allows accurate and accelerated phasing and imputation in a single tool by algorithmic and technical improvements and new features. We compared accuracy and runtime of EagleImp with Eagle2, PBWT and prominent imputation servers using whole-genome sequencing data from the 1000 Genomes Project, the Haplotype Reference Consortium and simulated data with 1 million reference genomes. EagleImp was 2-30 times faster (depending on the single or multiprocessor configuration selected and the size of the reference panel) than Eagle2 combined with PBWT, with the same or better phasing and imputation quality in all tested scenarios. For common variants investigated in typical genome-wide association studies, EagleImp provided same or higher imputation accuracy than the Sanger Imputation Service, Michigan Imputation Server and the newly developed TOPMed Imputation Server, despite larger (not publicly available) reference panels. Additional features include automated chromosome splitting and memory management at runtime to avoid job aborts, fast reading and writing of large files and various user-configurable algorithm and output options. Due to the technical optimizations, EagleImp can perform fast and accurate reference-based phasing and imputation and is ready for future large reference panels in the order of 1 million genomes. AVAILABILITY AND IMPLEMENTATION: EagleImp is implemented in C++ and freely available for download at https://github.com/ikmb/eagleimp. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The functional mechanisms underlying disease association remain unknown for Genome-wide Association Studies (GWAS) susceptibility variants located outside coding regions. Synthesis of effects from multiple surrounding functional variants has been suggested as an explanation of hard-to-interpret findings. We define filter criteria based on linkage disequilibrium measures and allele frequencies which reflect expected properties of synthesizing variant sets. For eligible candidate sets, we search for haplotype markers that are highly correlated with associated variants. RESULTS: Via simulations we assess the performance of our approach and suggest parameter settings which guarantee 95% sensitivity at 20-fold reduced computational cost. We apply our method to 1000 Genomes data and confirmed Crohn's Disease (CD) and Type 2 Diabetes (T2D) variants. A proportion of 36.9% allowed explanation by three-variant-haplotypes carrying at least two functional variants, as compared to 16.4% for random variants ([Formula: see text]). Association could be explained by missense variants for MUC19, PER3 (CD) and HMG20A (T2D). In a CD GWAS-imputed using haplotype reference consortium data (64 976 haplotypes)-we could confirm the syntheses of MUC19 and PER3 and identified synthesis by missense variants for 6 further genes (ZGPAZ, GPR65, CLN3/NPIPB8, LOC102723878, rs2872507, GCKR). In all instances, the odds ratios of the synthesizing haplotypes were virtually identical to that of the index SNP. In summary, we demonstrate the potential of synthesis analysis to guide functional follow-up of GWAS findings. AVAILABILITY AND IMPLEMENTATION: All methods are implemented in the C/C ++ toolkit GetSynth, available at http://sourceforge.net/projects/getsynth/ CONTACT: [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

High-throughput genotyping technologies (such as SNP-arrays) allow the rapid collection of up to a few million genetic markers of an individual. Detecting epistasis (based on 2-SNP interactions) in Genome-Wide Association Studies is an important but time consuming operation since statistical computations have to be performed for each pair of measured markers. Computational methods to detect epistasis therefore suffer from prohibitively long runtimes; e.g., processing a moderately-sized dataset consisting of about 500,000 SNPs and 5,000 samples requires several days using state-of-the-art tools on a standard 3 GHz CPU. In this paper, we demonstrate how this task can be accelerated using a combination of fine-grained and coarse-grained parallelism on two different computing systems. The first architecture is based on reconfigurable hardware (FPGAs) while the second architecture uses multiple GPUs connected to the same host. We show that both systems can achieve speedups of around four orders-of-magnitude compared to the sequential implementation. This significantly reduces the runtimes for detecting epistasis to only a few minutes for moderately-sized datasets and to a few hours for large-scale datasets.

BACKGROUND: Transposable elements are abundant in eukaryotic genomes and it is believed that they have a significant impact on the evolution of gene and chromosome structure. While there are several completed eukaryotic genome projects, there are only few high quality genome wide annotations of transposable elements. Therefore, there is a considerable demand for computational identification of transposable elements. LTR retrotransposons, an important subclass of transposable elements, are well suited for computational identification, as they contain long terminal repeats (LTRs). RESULTS: We have developed a software tool LTRharvest for the de novo detection of full length LTR retrotransposons in large sequence sets. LTRharvest efficiently delivers high quality annotations based on known LTR transposon features like length, distance, and sequence motifs. A quality validation of LTRharvest against a gold standard annotation for Saccharomyces cerevisae and Drosophila melanogaster shows a sensitivity of up to 90% and 97% and specificity of 100% and 72%, respectively. This is comparable or slightly better than annotations for previous software tools. The main advantage of LTRharvest over previous tools is (a) its ability to efficiently handle large datasets from finished or unfinished genome projects, (b) its flexibility in incorporating known sequence features into the prediction, and (c) its availability as an open source software. CONCLUSION: LTRharvest is an efficient software tool delivering high quality annotation of LTR retrotransposons. It can, for example, process the largest human chromosome in approx. 8 minutes on a Linux PC with 4 GB of memory. Its flexibility and small space and run-time requirements makes LTRharvest a very competitive candidate for future LTR retrotransposon annotation projects. Moreover, the structured design and implementation and the availability as open source provides an excellent base for incorporating novel concepts to further improve prediction of LTR retrotransposons.