Search

BackgroundImaging ofin vitroneuronal differentiation and measurements of cell morphologies has led to novel insights into neuronal development. Live-cell imaging techniques and large datasets of images has increased the demand for automated pipelines for quantitative analysis of neuronal morphological metrics.ResultsWe present ANDA, an analysis workflow for quantification of various aspects of neuronal morphology from high-throughput live-cell imaging screens. This tool automates the analysis of neuronal cell numbers, neurite lengths and neurite attachment points. We used rat, chicken and humanin vitromodels for neuronal differentiation and have demonstrated the accuracy, versatility, and efficiency of the tool.ConclusionsANDA is an open-source tool that is easy to use and capable of automated processing from time-course measurements of neuronal cells. The strength of this pipeline is the capability to analyse high-throughput imaging screens.

As decreasing DNA sequencing costs leads to a steadily increasing rate of generated data, the development of efficient algorithms for processing of the sequence data is increasingly important to reduce costs and energy consumption. Recent work have shown that genotyping can be done efficiently and accurately using alignment-free methods that are based on analyzing kmers from sequenced reads. In particular, we have recently presented the KAGE genotyper, which uses an efficient pangenome representation of known individuals in a population to further increase accuracy and efficiency. While existing genotypers like KAGE use the Central Processing Unit (CPU) to count and analyze kmers, the Graphical Processing Unit (GPU) has shown promising for reducing runtime for similar problems.We here present GKAGE, a new and improved version of KAGE that utilizes the GPU to further increase the computational efficiency. This is done by counting and analyzing large amounts of kmers in the many parallel cores of a GPU. We show that GKAGE is, on hardware of comparable cost, able to genotype an individual up to an order of magnitude faster than KAGE while producing the same output, which makes it by far the fastest genotyper available today. GKAGE can run on consumer-grade GPUs, and enables genotyping of a human sample in only a matter of minutes without the need for expensive high-performance computers. GKAGE is open source and available athttps://github.com/kage-genotyper/kage.

Python is a popular and widespread programming language for scientific computing, in large part due to the powerfularray programminglibrary NumPy, which makes it easy to write clean, vectorized and efficient code for handling large datasets. A challenge with using array programming for biological data is that the data is often non-numeric and variable-length (such as DNA sequences), inhibiting out-of-the-box use of standard array programming techniques. Thus, a tradition in bioinformatics has been to use low-level languages like C and C++ to write efficient code. This makes the tools less transparent to the average computational biologist - making them harder to understand, modify and contribute to.We here present a new Python package BioNumPy, which adds a layer on top of NumPy in order to enable intuitive array programming on biological datasets. BioNumPy is able to efficiently load biological datasets (e.g. FASTQ-files, BED-files and BAM-files) into NumPy-like data structures, so that NumPy operations like indexing, vectorized functions and reductions can be applied to the data. We show that BioNumPy is considerably faster than vanilla Python and other Python packages for common bioinformatics tasks, and in many cases as fast as tools written in C/C++. BioNumPy thus bridges a long-lasting gap in bioinformatics, allowing the same programming language (Python) to be used across the full spectrum from quick and simple scripts to computationally efficient processing of large-scale data.

SummaryAdaptive immune receptor (AIR) repertoires (AIRRs) record past immune encounters with exquisite specificity. Therefore, identifying identical or similar AIR sequences across individuals is a key step in AIRR analysis for revealing convergent immune response patterns that may be exploited for diagnostics and therapy. Existing methods for quantifying AIRR overlap do not scale with increasing dataset numbers and sizes. To address this limitation, we developed CompAIRR, which enables ultra-fast computation of AIRR overlap, based on either exact or approximate sequence matching. CompAIRR improves computational speed 1000-fold relative to the state of the art and uses only one-third of the memory: on the same machine, the exact pairwise AIRR overlap of 104 AIRRs with 105 sequences is found in ∼17 minutes, while the fastest alternative tool requires 10 days. CompAIRR has been integrated with the machine learning ecosystem immuneML to speed up various commonly used AIRR-based machine learning applications.Availability and implementationCompAIRR code and documentation are available at https://github.com/uio-bmi/compairr. Docker images are available at https://hub.docker.com/r/torognes/compairr. The scripts used for benchmarking and creating figures, and all raw data, may be found at https://github.com/uio-bmi/compairr-benchmarking.

Generative machine learning (ML) has been postulated to be a major driver in the computational design of antigen-specific monoclonal antibodies (mAb). However, efforts to confirm this hypothesis have been hindered by the infeasibility of testing arbitrarily large numbers of antibody sequences for their most critical design parameters: paratope, epitope, affinity, and developability. To address this challenge, we leveraged a lattice-based antibody-antigen binding simulation framework, which incorporates a wide range of physiological antibody binding parameters. The simulation framework enables both the computation of antibody-antigen 3D-structures as well as functions as an oracle for unrestricted prospective evaluation of the antigen specificity of ML-generated antibody sequences. We found that a deep generative model, trained exclusively on antibody sequence (1D) data can be used to design native-like conformational (3D) epitope-specific antibodies, matching or exceeding the training dataset in affinity and developability variety. Furthermore, we show that transfer learning enables the generation of high-affinity antibody sequences from low-N training data. Finally, we validated that the antibody design insight gained from simulated antibody-antigen binding data is applicable to experimental real-world data. Our work establishes a priori feasibility and the theoretical foundation of high-throughput ML-based mAb design.HighlightsA large-scale dataset of 70M [3 orders of magnitude larger than the current state of the art] synthetic antibody-antigen complexes, that reflect biological complexity, allows the prospective evaluation of antibody generative deep learningCombination of generative learning, synthetic antibody-antigen binding data, and prospective evaluation shows that deep learning driven antibody design and discovery at an unconstrained level is feasibleTransfer learning (low-N learning) coupled to generative learning shows that antibody-binding rules may be transferred across unrelated antibody-antigen complexesExperimental validation of antibody-design conclusions drawn from deep learning on synthetic antibody-antigen binding dataGraphical abstractWe leverage large synthetic ground-truth data to demonstrate the (A,B) unconstrained deep generative learning-based generation of native-like antibody sequences, (C) the prospective evaluation of conformational (3D) affinity, paratope-epitope pairs, and developability. (D) Finally, we show increased generation quality of low-N-based machine learning models via transfer learning.

Machine learning (ML) is a key technology to enable accurate prediction of antibody-antigen binding, a prerequisite for in silico vaccine and antibody design. Two orthogonal problems hinder the current application of ML to antibody-specificity prediction and the benchmarking thereof: (i) The lack of a unified formalized mapping of immunological antibody specificity prediction problems into ML notation and (ii) the unavailability of large-scale training datasets. Here, we developed the Absolut! software suite that allows the parameter-based unconstrained generation of synthetic lattice-based 3D-antibody-antigen binding structures with ground-truth access to conformational paratope, epitope, and affinity. We show that Absolut!-generated datasets recapitulate critical biological sequence and structural features that render antibody-antigen binding prediction challenging. To demonstrate the immediate, high-throughput, and large-scale applicability of Absolut!, we have created an online database of 1 billion antibody-antigen structures, the extension of which is only constrained by moderate computational resources. We translated immunological antibody specificity prediction problems into ML tasks and used our database to investigate paratope-epitope binding prediction accuracy as a function of structural information encoding, dataset size, and ML method, which is unfeasible with existing experimental data. Furthermore, we found that in silico investigated conditions, predicted to increase antibody specificity prediction accuracy, align with and extend conclusions drawn from experimental antibody-antigen structural data. In summary, the Absolut! framework enables the development and benchmarking of ML strategies for biotherapeutics discovery and design. O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=194 SRC="FIGDIR/small/451258v2_ufig1.gif" ALT="Figure 1"> View larger version (81K): org.highwire.dtl.DTLVardef@418bfforg.highwire.dtl.DTLVardef@14e1ec9org.highwire.dtl.DTLVardef@1c75b50org.highwire.dtl.DTLVardef@13bf69d_HPS_FORMAT_FIGEXP M_FIG O_FLOATNOGraphical abstractC_FLOATNO The software framework Absolut! enables (A,B) the generation of virtually arbitrarily large numbers of in silico 3D-antibody-antigen structures, (C,D) the formalization of antibody specificity as machine learning (ML) tasks as well as the exploration of ML strategies for paratope-epitope prediction. C_FIG Highlights- Software framework Absolut! to generate an arbitrarily large number of in silico 3D-antibody-antigen structures - Generation of one billion in silico antigen-antibody structures reflecting biological layers of complexity that make ML predictions challenging - Immunological antibody specificity prediction problems formalized as machine learning tasks for which the in silico complexes are immediately usable as benchmark. - Exploration of machine learning architectures for paratope-epitope interaction prediction accuracy as a function of neural network depth, dataset size, and sequence-structure encoding

1Gene co-expression measurements are widely used in computational biology to identify coordinated expression patterns across a group of samples, which may indicate that these genes are controlled by the same transcriptional regulatory program, or involved in common biological processes. Gene co-expression is generally estimated from RNA-Seq data, which are generally normalized to remove technical variability. Here, we find and demonstrate that certain normalization methods, in particular quantile-based methods, can introduce false-positive associations between genes, and that this can consequently hamper downstream co-expression network analysis. Quantile-based normalization can, however, be extremely powerful. In particular when preprocessing large-scale heterogeneous data, quantile-based normalization can be applied to remove technical variability while maintaining global differences in expression for samples with different biological attributes. We therefore developed CAIMAN, a method to correct for false-positive associations that may arise from normalization of RNA-Seq data. CAIMAN utilizes a Gaussian mixture model to fit the distribution of gene expression and to adaptively select the threshold to define lowly expressed genes, which are prone to form false-positive associations. Thereafter, CAIMAN corrects the normalized expression for these genes by removing the variability across samples that might lead to false-positive associations. Moreover, CAIMAN avoids arbitrary gene filtering and retains associations to genes that only express in small subgroups of samples, highlighting its potential future impact on network modeling and other association-based approaches in large-scale heterogeneous data.

Adaptive immune receptor repertoires (AIRR) are key targets for biomedical research as they record past and ongoing adaptive immune responses. The capacity of machine learning (ML) to identify complex discriminative sequence patterns renders it an ideal approach for AIRR-based diagnostic and therapeutic discovery. To date, widespread adoption of AIRR ML has been inhibited by a lack of reproducibility, transparency, and interoperability. immuneML (immuneml.uio.no) addresses these concerns by implementing each step of the AIRR ML process in an extensible, open-source software ecosystem that is based on fully specified and shareable workflows. To facilitate widespread user adoption, immuneML is available as a command-line tool and through an intuitive Galaxy web interface, and extensive documentation of workflows is provided. We demonstrate the broad applicability of immuneML by (i) reproducing a large-scale study on immune state prediction, (ii) developing, integrating, and applying a novel method for antigen specificity prediction, and (iii) showcasing streamlined interpretability-focused benchmarking of AIRR ML.

A central mechanism in machine learning is to identify, store, and recognize patterns. How to learn, access, and retrieve such patterns is crucial in Hopfield networks and the more recent transformer architectures. We show that the attention mechanism of transformer architectures is actually the update rule of modern Hop-field networks that can store exponentially many patterns. We exploit this high storage capacity of modern Hopfield networks to solve a challenging multiple instance learning (MIL) problem in computational biology: immune repertoire classification. Accurate and interpretable machine learning methods solving this problem could pave the way towards new vaccines and therapies, which is currently a very relevant research topic intensified by the COVID-19 crisis. Immune repertoire classification based on the vast number of immunosequences of an individual is a MIL problem with an unprecedentedly massive number of instances, two orders of magnitude larger than currently considered problems, and with an extremely low witness rate. In this work, we present our novel method DeepRC that integrates transformer-like attention, or equivalently modern Hopfield networks, into deep learning architectures for massive MIL such as immune repertoire classification. We demonstrate that DeepRC outperforms all other methods with respect to predictive performance on large-scale experiments, including simulated and real-world virus infection data, and enables the extraction of sequence motifs that are connected to a given disease class. Source code and datasets: https://github.com/ml-jku/DeepRC

Graph-based reference genomes have become popular as they allow read mapping and follow-up analyses in settings where the exact haplotypes underlying a high-throughput sequencing experiment are not precisely known. Two recent papers show that mapping to graph-based reference genomes can improve accuracy as compared to methods using linear references. Both of these methods index the sequences for most paths up to a certain length in the graph in order to enable direct mapping of reads containing common variants. However, the combinatorial explosion of possible paths through nearby variants also leads to a huge search space and an increased chance of false positive alignments to highly variable regions.We here assess two prominent graph-based read mappers against a novel hybrid baseline approach that combines an initial path determination with a tuned linear read mapping method. We show, using a previously proposed benchmark, that this simple approach is able to improve accuracy of read-mapping to graph-based reference genomes.Our method is implemented in a tool, Two-step Graph Mapper, which is available at https://github.com/uio-bmi/two_step_graph_mapper along with data and scripts for reproducing the experiments.

BackgroundThe generation and systematic collection of genome-wide data is ever-increasing. This vast amount of data has enabled researchers to study relations between a variety of genomic and epigenomic features, including genetic variation, gene regulation, and phenotypic traits. Such relations are typically investigated by comparatively assessing genomic co-occurrence. Technically, this corresponds to assessing the similarity of pairs of genome-wide binary vectors. A variety of metrics have been proposed for this problem in other fields like ecology. However, while several of these metrics have been employed for assessing genomic co-occurrence, their appropriateness for the genomic setting has never been investigated.ResultsWe show that the choice of metric may strongly influence results and propose two alternative modelling assumptions that can be used to guide this choice. On both simulated and real genomic data, the Jaccard index is strongly affected by dataset size and should be used with caution. The Forbes coefficient (fold change) and tetrachoric correlation are less affected by dataset size, but one should be aware of increased variance for small datasets.AvailabilityAll results on simulated and real data can be inspected and reproduced athttps://hyperbrowser.uio.no/sim-measure

BackgroundThe difficulties associated with sequencing and assembling some regions of the DNA sequence result in gaps in the reference genomes that are typically represented as stretches of Ns. Although the presence of assembly gaps causes a slight reduction in the mapping rate in many experimental settings, that does not invalidate the typical statistical testing comparing read count distributions across experimental conditions. However, we hypothesize that not handling assembly gaps in the null model may confound statistical testing of co-localization of genomic features.ResultsFirst, we performed a series of explorative analyses to understand whether and how the public genomic tracks intersect the assembly gaps track (hg19). The findings rightly confirm that the genomic regions in public genomic tracks intersect very little with assembly gaps and the intersection was observed only at the beginning and end regions of the assembly gaps rather than covering the whole gap sizes. Further, we simulated a set of query and reference genomic tracks in a way that nullified any dependence between them to test our hypothesis that not avoiding assembly gaps in the null model would result in spurious inflation of statistical significance. We then contrasted the distributions of test statistics and p-values of Monte Carlo simulation-based permutation tests that either avoided or not avoided assembly gaps in the null model when testing for significant co-localization between a pair of query and reference tracks. We observed that the statistical tests that did not account for the assembly gaps in the null model resulted in a distribution of the test statistic that is shifted to the right and a distribu tion of p-values that is shifted to the left (leading to inflated significance).ConclusionOur results shows that not accounting for assembly gaps in statistical testing of co-localization analysis may lead to false positives and over-optimistic findings.

Genomic locations are represented as coordinates on a specific genome build version, but the build information is frequently missing when coordinates are provided. It is essential to correctly interpret and analyse the genomic intervals contained in genomic track files. Here, we demonstrate that this crucial metadatum (or rather datum) is often isolated from the genomic track files in public repositories and journal articles, which could be a major time thief. We propose best practices to ensure that genome build version is always carried along with genomic track files. Although not a substitute to the best practices, we also provide a tool to predict the genome build version of genomic track files.

Genome-wide, cell-type-specific profiles are being systematically generated for numerous genomic and epigenomic features. There is, however, no universally applicable analytical methodology for such data. We present GSuite HyperBrowser, the first comprehensive solution for integrative analysis of dataset collections across the genome and epigenome. The GSuite HyperBrowser is an open-source system for streamlined acquisition and customizable statistical analysis of large collections of genome-wide datasets. The system is based on new computational and statistical methodologies that permit comparative and confirmatory analyses across multiple disparate data sources. Expert guidance and reproducibility are facilitated via a Galaxy-based web-interface. The software is available athttps://hyperbrowser.uio.no/gsuite