MOTIVATION: In this work, we present an analytical method for quantifying both single-cell morphologies and cell network topologies of tumor cell populations and use it to predict 3D cell behavior. RESULTS: We utilized a supervised deep learning approach to perform instance segmentation on label-free live cell images across a wide range of cell densities. We measured cell shape properties and characterized network topologies for 136 single-cell clones (SCCs) derived from the YUMM1.7 and YUMMER1.7 mouse melanoma cell lines. Using an unsupervised clustering algorithm, we identified six distinct morphological subclasses. We further observed differences in tumor growth and invasion dynamics across subclasses in an in vitro 3D spheroid model. Compared to existing methods for quantifying 2D or 3D phenotype, our analytical method requires less time, needs no specialized equipment, and is capable of much higher throughput, making it ideal for applications such as high-throughput drug screening and clinical diagnosis. AVAILABILITY AND IMPLEMENTATION: //github.com/trevor-chan/Melanoma_NetworkMorphology. SUPPLEMENTAL INFORMATION: Bioinformatics online.

SUMMARY: Recent innovations in single-cell chromatin accessibility sequencing (scCAS) have revolutionized the characterization of epigenomic heterogeneity. Estimation of the number of cell types is a crucial step for downstream analyses and biological implications. However, efforts to perform estimation specifically for scCAS data are limited. Here we propose ASTER, an ensemble learning-based tool for accurately estimating the number of cell types in scCAS data. ASTER outperformed baseline methods in systematic evaluation on 27 datasets of various protocols, sizes, numbers of cell types, degrees of cell-type imbalance, cell states, and qualities, providing valuable guidance for scCAS data analysis. AVAILABILITY AND IMPLEMENTATION: ASTER along with detailed documentation is freely accessible at https://aster.readthedocs.io/ under the MIT License. It can be seamlessly integrated into existing scCAS analysis workflows. The source code is available at https://github.com/biox-nku/aster. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: RNA viruses tend to mutate constantly. While many of the variants are neutral, some can lead to higher transmissibility or virulence. Accurate assembly of complete viral genomes enables identification of underlying variants, which are essential for studying virus evolution and elucidating the relationship between genotypes and virus properties. Recently, third-generation sequencing platforms such as Nanopore sequencers have been used for real-time virus sequencing for Ebola, Zika, COVID-19, etc. However, their high per-base error rate prevents accurate reconstruction of the viral genome. RESULTS: In this work, we introduce a new tool, AccuVIR, for viral genome assembly and polishing using error-prone long reads. It can better distinguish sequencing errors from true variants based on the key observation that sequencing errors can disrupt the gene structures of viruses, which usually have high density of coding regions. Our experimental results on both simulated and real third-generation sequencing data demonstrated its superior performance on generating more accurate viral genomes than generic assembly or polish tools. AVAILABILITY: The source code and the documentation of AccuVIR are available at https://github. com/rainyrubyzhou/AccuVIR. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: In many modern bioinformatics applications, such as statistical genetics, or single-cell analysis, one frequently encounters datasets which are orders of magnitude too large for conventional in-memory analysis. To tackle this challenge, we introduce SIMBSIG, a highly scalable Python package which provides a scikit-learn-like interface for out-of-core, GPU-enabled similarity searches, principal component analysis, and clustering. Due to the PyTorch backend it is highly modular and particularly tailored to many data types with a particular focus on biobank data analysis. AVAILABILITY: SIMBSIG is freely available from PyPI and its source code and documentation can be found on GitHub (https://github.com/BorgwardtLab/simbsig) under a BSD-3 license.

MOTIVATION: Spatially resolved gene expression profiles are the key to exploring the cell type spatial distributions and understanding the architecture of tissues. Many spatially resolved transcriptomics (SRT) techniques do not provide single-cell resolutions, but they measure gene expression profiles on captured locations (spots) instead, which are mixtures of potentially heterogeneous cell types. Currently, several cell type deconvolution methods have been proposed to deconvolute SRT data. Due to the different model strategies of these methods, their deconvolution results also vary. RESULTS: Leveraging the strengths of multiple deconvolution methods, we introduce a new weighted ensemble learning deconvolution method, EnDecon, to predict cell type compositions on SRT data in this work. EnDecon integrates multiple base deconvolution results using a weighted optimization model to generate a more accurate result. Simulation studies demonstrate that EnDecon outperforms the competing methods and the learned weights assigned to base deconvolution methods have high positive correlations with the performances of these base methods. Applied to real datasets from different spatial techniques, EnDecon identifies multiple cell types on spots, localizes these cell types to specific spatial regions, and distinguishes distinct spatial colocalization and enrichment patterns, providing valuable insights into spatial heterogeneity and regionalization of tissues. AVAILABILITY: The source code is available at https://github.com/Zhangxf-ccnu/EnDecon. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: The effectiveness of CRISPR/Cas9-mediated genome editing experiments largely depends on the guide RNA (gRNA) used by the CRISPR/Cas9 system for target recognition and cleavage activation. Careful design is necessary to select a gRNA with high editing efficiency at the on-target site and with minimum off-target potential. Here, we present our webserver for gRNA design with a user-friendly graphical interface, which provides interoperability between our on- and off-target prediction tools, CRISPRon and CRISPRoff, for a complete and streamlined gRNA selection. AVAILABILITY AND IMPLEMENTATION: The graphical interface uses the Integrative Genomic Viewer (IGV) JavaScript plugin. The backend tools are implemented in Python and C. The CRISPRon and CRISPRoff webservers and command-line tools are freely available at https://rth.dk/resources/crispr.

MOTIVATION: A global medical crisis like the coronavirus disease 2019 (COVID-19) pandemic requires interdisciplinary and highly collaborative research from all over the world. One of the key challenges for collaborative research is a lack of interoperability among various heterogeneous data sources. Interoperability, standardization and mapping of datasets are necessary for data analysis and applications in advanced algorithms such as developing personalized risk prediction modeling. RESULTS: To ensure the interoperability and compatibility among COVID-19 datasets, we present here a common data model (CDM) which has been built from 11 different COVID-19 datasets from various geographical locations. The current version of the CDM holds 4639 data variables related to COVID-19 such as basic patient information (age, biological sex and diagnosis) as well as disease-specific data variables, for example, Anosmia and Dyspnea. Each of the data variables in the data model is associated with specific data types, variable mappings, value ranges, data units and data encodings that could be used for standardizing any dataset. Moreover, the compatibility with established data standards like OMOP and FHIR makes the CDM a well-designed CDM for COVID-19 data interoperability. AVAILABILITY AND IMPLEMENTATION: The CDM is available in a public repo here: https://github.com/Fraunhofer-SCAI-Applied-Semantics/COVID-19-Global-Model. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: ADViSELipidomics is a novel Shiny app for preprocessing, analyzing and visualizing lipidomics data. It handles the outputs from LipidSearch and LIQUID for lipid identification and quantification and the data from the Metabolomics Workbench. ADViSELipidomics extracts information by parsing lipid species (using LIPID MAPS classification) and, together with information available on the samples, performs several exploratory and statistical analyses. When the experiment includes internal lipid standards, ADViSELipidomics can normalize the data matrix, providing normalized concentration values per lipids and samples. Moreover, it identifies differentially abundant lipids in simple and complex experimental designs, dealing with batch effect correction. Finally, ADViSELipidomics has a user-friendly graphical user interface and supports an extensive series of interactive graphics. AVAILABILITY AND IMPLEMENTATION: ADViSELipidomics is freely available at https://github.com/ShinyFabio/ADViSELipidomics. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Genome wide association studies elucidate links between genotypes and phenotypes. Recent studies point out the interest of conducting such experiments using k-mers as the base signal instead of single-nucleotide polymorphisms. We propose a tool, kmdiff, that performs differential k-mer analyses on large sequencing cohorts in an order of magnitude less time and memory than previously possible. AVAILABILITYAND IMPLEMENTATION: https://github.com/tlemane/kmdiff. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Large-scale neuronal morphologies are essential to neuronal typing, connectivity characterization and brain modeling. It is widely accepted that automation is critical to the production of neuronal morphology. Despite previous survey papers about neuron tracing from light microscopy data in the last decade, thanks to the rapid development of the field, there is a need to update recent progress in a review focusing on new methods and remarkable applications. RESULTS: This review outlines neuron tracing in various scenarios with the goal to help the community understand and navigate tools and resources. We describe the status, examples and accessibility of automatic neuron tracing. We survey recent advances of the increasingly popular deep-learning enhanced methods. We highlight the semi-automatic methods for single neuron tracing of mammalian whole brains as well as the resulting datasets, each containing thousands of full neuron morphologies. Finally, we exemplify the commonly used datasets and metrics for neuron tracing bench testing.

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: Single-molecule localization microscopy resolves individual fluorophores or fluorescence-labeled biomolecules. Data are provided as a set of localizations that distribute normally around the true fluorophore position with a variance determined by the localization precision. Characterizing the spatial fluorophore distribution to differentiate between resolution-limited localization clusters, which resemble individual biomolecules, and extended structures, which represent aggregated molecular complexes, is a common challenge. RESULTS: We demonstrate the use of the convex hull and related hull properties of localization clusters for diagnostic purposes, as a parameter for cluster selection or as a tool to determine localization precision. AVAILABILITY AND IMPLEMENTATION: https://github.com/super-resolution/Ebert-et-al-2022-supplement. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: EvAM-Tools is an R package and web application that provides a unified interface to state-of-the-art cancer progression models and, more generally, evolutionary models of event accumulation. The output includes, in addition to the fitted models, the transition (and transition rate) matrices between genotypes and the probabilities of evolutionary paths. Generation of random cancer progression models is also available. Using the GUI in the web application, users can easily construct models (modifying directed acyclic graphs of restrictions, matrices of mutual hazards or specifying genotype composition), generate data from them (with user-specified observational/genotyping error) and analyze the data. AVAILABILITY AND IMPLEMENTATION: Implemented in R and C; open source code available under the GNU Affero General Public License v3.0 at https://github.com/rdiaz02/EvAM-Tools. Docker images freely available from https://hub.docker.com/u/rdiaz02. Web app freely accessible at https://iib.uam.es/evamtools. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The 'glycoEnzymes' include a set of proteins having related enzymatic, metabolic, transport, structural and cofactor functions. Currently, there is no established ontology to describe glycoEnzyme properties and to relate them to glycan biosynthesis pathways. RESULTS: We present GlycoEnzOnto, an ontology describing 403 human glycoEnzymes curated along 139 glycosylation pathways, 134 molecular functions and 22 cellular compartments. The pathways described regulate nucleotide-sugar metabolism, glycosyl-substrate/donor transport, glycan biosynthesis and degradation. The role of each enzyme in the glycosylation initiation, elongation/branching and capping/termination phases is described. IUPAC linear strings present systematic human/machine-readable descriptions of individual reaction steps and enable automated knowledge-based curation of biochemical networks. All GlycoEnzOnto knowledge is integrated with the Gene Ontology biological processes. GlycoEnzOnto enables improved transcript overrepresentation analyses and glycosylation pathway identification compared to other available schema, e.g. KEGG and Reactome. Overall, GlycoEnzOnto represents a holistic glycoinformatics resource for systems-level analyses. AVAILABILITY AND IMPLEMENTATION: https://github.com/neel-lab/GlycoEnzOnto. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: The detection of the structural variants (SVs) using Illumina sequencing of human DNA is not an easy task. Multiple approaches have been proposed; however, all the methods have their limitations. In this article, we present ConsensuSV pipeline that aids the research in complex variant detection. By using consensus meta-approach, eight independent SV callers are being used to identify a uniform set of high-quality SVs. The pipeline works using raw sequencing data and performs all the necessary steps automatically, significantly reducing the researchers' time required for processing the data. The output files contain SVs, single nucleotide polymorphisms and Indels. The pipeline uses luigi framework, allowing the software to be run efficiently and parallelly using the high-performance computing infrastructure. We strongly believe that the software is useful to the scientific community interested in the germline variant detection. AVAILABILITY AND IMPLEMENTATION: https://github.com/SFGLab/ConsensuSV-pipeline. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Interactions between proteins help us understand how genes are functionally related and how they contribute to phenotypes. Experiments provide imperfect 'ground truth' information about a small subset of potential interactions in a specific biological context, which can then be extended to the whole genome across different contexts, such as conditions, tissues or species, through machine learning methods. However, evaluating the performance of these methods remains a critical challenge. Here, we propose to evaluate the generalizability of gene characterizations through the shape of performance curves. RESULTS: We identify Functional Equivalence Classes (FECs), subsets of annotated and unannotated genes that jointly drive performance, by assessing the presence of straight lines in ROC curves built from gene-centric prediction tasks, such as function or interaction predictions. FECs are widespread across data types and methods, they can be used to evaluate the extent and context-specificity of functional annotations in a data-driven manner. For example, FECs suggest that B cell markers can be decomposed into shared primary markers (10-50 genes), and tissue-specific secondary markers (100-500 genes). In addition, FECs suggest the existence of functional modules that span a wide range of the genome, with marker sets spanning at most 5% of the genome and data-driven extensions of Gene Ontology sets spanning up to 40% of the genome. Simple to assess visually and statistically, the identification of FECs in performance curves paves the way for novel functional characterization and increased robustness in the definition of functional gene sets. AVAILABILITY AND IMPLEMENTATION: Code for analyses and figures is available at https://github.com/yexilein/pyroc. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Drug-drug interaction (DDI) prediction is a challenging problem in pharmacology and clinical applications. With the increasing availability of large biomedical databases, large-scale biological knowledge graphs containing drug information have been widely used for DDI prediction. However, large knowledge graphs inevitably suffer from data noise problems, which limit the performance and interpretability of models based on the knowledge graph. Recent studies attempt to improve models by introducing inductive bias through an attention mechanism. However, they all only depend on the topology of entity nodes independently to generate fixed attention pathways, without considering the semantic diversity of entity nodes in different drug pair links. This makes it difficult for models to select more meaningful nodes to overcome data quality limitations and make more interpretable predictions. RESULTS: To address this issue, we propose a Link-aware Graph Attention method for DDI prediction, called LaGAT, which is able to generate different attention pathways for drug entities based on different drug pair links. For a drug pair link, the LaGAT uses the embedding representation of one of the drugs as a query vector to calculate the attention weights, thereby selecting the appropriate topological neighbor nodes to obtain the semantic information of the other drug. We separately conduct experiments on binary and multi-class classification and visualize the attention pathways generated by the model. The results prove that LaGAT can better capture semantic relationships and achieves remarkably superior performance over both the classical and state-of-the-art models on DDI prediction. AVAILABILITYAND IMPLEMENTATION: The source code and data are available at https://github.com/Azra3lzz/LaGAT. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Current approaches detect conserved genomic order either at chromosomal (macrosynteny) or at subchromosomal scales (microsynteny). The latter generally requires collinearity and hard thresholds on syntenic region size, thus excluding a major proportion of syntenies with recent expansions or minor rearrangements. 'SYNPHONI' bridges the gap between micro- and macrosynteny detection, providing detailed information on both synteny conservation and transformation throughout the evolutionary history of animal genomes. AVAILABILITY AND IMPLEMENTATION: Source code is freely available at https://github.com/nsmro/SYNPHONI, implemented in Python 3.9. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Low-complexity domains (LCDs) in proteins are regions enriched in a small subset of amino acids. LCDs exist in all domains of life, often have unusual biophysical behavior, and function in both normal and pathological processes. We recently developed an algorithm to identify LCDs based predominantly on amino acid composition thresholds. Here, we have integrated this algorithm with a webserver and augmented it with additional analysis options. Specifically, users can (i) search for LCDs in whole proteomes by setting minimum composition thresholds for individual or grouped amino acids, (ii) submit a known LCD sequence to search for similar LCDs, (iii) search for and plot LCDs within a single protein, (iv) statistically test for enrichment of LCDs within a user-provided protein set and (v) specifically identify proteins with multiple types of LCDs. AVAILABILITY AND IMPLEMENTATION: The LCD-Composer server can be accessed at http://lcd-composer.bmb.colostate.edu. The corresponding command-line scripts can be accessed at https://github.com/RossLabCSU/LCD-Composer/tree/master/WebserverScripts.

MOTIVATION: Haplotypes are the set of alleles co-occurring on a single chromosome and inherited together to the next generation. Because a monoploid reference genome loses this co-occurrence information, it has limited use in associating phenotypes with allelic combinations of genotypes. Therefore, methods to reconstruct the complete haplotypes from DNA sequencing data are crucial. Recently, several attempts have been made at haplotype reconstructions, but significant limitations remain. High-quality continuous haplotypes cannot be created reliably, particularly when there are few differences between the homologous chromosomes. RESULTS: Here, we introduce HAT, a haplotype assembly tool that exploits short and long reads along with a reference genome to reconstruct haplotypes. HAT tries to take advantage of the accuracy of short reads and the length of the long reads to reconstruct haplotypes. We tested HAT on the aneuploid yeast strain Saccharomyces pastorianus CBS1483 and multiple simulated polyploid datasets of the same strain, showing that it outperforms existing tools. AVAILABILITY AND IMPLEMENTATION: https://github.com/AbeelLab/hat/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: The cellular system of a living organism is composed of interacting bio-molecules that control cellular processes at multiple levels. Their correspondences are represented by tightly regulated molecular networks. The increase of omics technologies has favored the generation of large-scale disparate data and the consequent demand for simultaneously using molecular and functional interaction networks: gene co-expression, protein-protein interaction (PPI), genetic interaction and metabolic networks. They are rich sources of information at different molecular levels, and their effective integration is essential to understand cell functioning and their building blocks (proteins). Therefore, it is necessary to obtain informative representations of proteins and their proximity, that are not fully captured by features extracted directly from a single informational level. We propose BraneMF, a novel random walk-based matrix factorization method for learning node representation in a multilayer network, with application to omics data integration. RESULTS: We test BraneMF with PPI networks of Saccharomyces cerevisiae, a well-studied yeast model organism. We demonstrate the applicability of the learned features for essential multi-omics inference tasks: clustering, function and PPI prediction. We compare it to the state-of-the-art integration methods for multilayer networks. BraneMF outperforms baseline methods by achieving high prediction scores for a variety of downstream tasks. The robustness of results is assessed by an extensive parameter sensitivity analysis. AVAILABILITY AND IMPLEMENTATION: BraneMF's code is freely available at: https://github.com/Surabhivj/BraneMF, along with datasets, embeddings and result files. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Recent studies suggest that the loop extrusion activity of Structural Maintenance of Chromosomes complexes is central to proper organization of genomes in vivo. Polymer physics-based modeling of chromosome structure has been instrumental to assess which structures such extrusion can create. Only few laboratories however have the technical and computational expertise to create in silico models combining dynamic features of chromatin and loop extruders. Here, we present 3DPolyS-LE, a self-contained, easy to use modeling and simulation framework allowing non-specialists to ask how specific properties of loop extruders and boundary elements impact on 3D chromosome structure. 3DPolyS-LE also provides algorithms to compare predictions with experimental Hi-C data. AVAILABILITY AND IMPLEMENTATION: Software available at https://gitlab.com/togop/3DPolyS-LE; implemented in Python and Fortran 2003 and supported on any Unix-based operating system (Linux and Mac OS). SUPPLEMENTARY INFORMATION: Supplementary information are available at Bioinformatics online.

MOTIVATION: Lacking strict proofreading mechanisms, many RNA viruses can generate progeny with slightly changed genomes. Being able to characterize highly similar genomes (i.e. haplotypes) in one virus population helps study the viruses' evolution and their interactions with the host/other microbes. High-throughput sequencing data has become the major source for characterizing viral populations. However, the inherent limitation on read length by next-generation sequencing makes complete haplotype reconstruction difficult. RESULTS: In this work, we present a new tool named HaploDMF that can construct complete haplotypes using third-generation sequencing (TGS) data. HaploDMF utilizes a deep matrix factorization model with an adapted loss function to learn latent features from aligned reads automatically. The latent features are then used to cluster reads of the same haplotype. Unlike existing tools whose performance can be affected by the overlap size between reads, HaploDMF is able to achieve highly robust performance on data with different coverage, haplotype number and error rates. In particular, it can generate more complete haplotypes even when the sequencing coverage drops in the middle. We benchmark HaploDMF against the state-of-the-art tools on simulated and real sequencing TGS data on different viruses. The results show that HaploDMF competes favorably against all others. AVAILABILITY AND IMPLEMENTATION: The source code and the documentation of HaploDMF are available at https://github.com/dhcai21/HaploDMF. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Years of time-series gene expression studies have built a strong understanding of clock-controlled pathways across species. However, comparatively little is known about how 'non-clock' pathways influence clock function. We need a strong understanding of clock-coupled pathways in human tissues to better appreciate the links between disease and clock function. RESULTS: We developed a new computational approach to explore candidate pathways coupled to the clock in human tissues. This method, termed LTM, is an in silico screen to infer genetic influences on circadian clock function. LTM uses natural variation in gene expression in human data and directly links gene expression variation to clock strength independent of longitudinal data. We applied LTM to three human skin and one melanoma datasets and found that the cell cycle is the top candidate clock-coupled pathway in healthy skin. In addition, we applied LTM to thousands of tumor samples from 11 cancer types in the TCGA database and found that extracellular matrix organization-related pathways are tightly associated with the clock strength in humans. Further analysis shows that clock strength in tumor samples is correlated with the proportion of cancer-associated fibroblasts and endothelial cells. Therefore, we show both the power of LTM in predicting clock-coupled pathways and classify factors associated with clock strength in human tissues. AVAILABILITY AND IMPLEMENTATION: LTM is available on GitHub (https://github.com/gangwug/LTMR) and figshare (https://figshare.com/articles/software/LTMR/21217604) to facilitate its use. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: Understanding the mechanism of action of a protein or designing better ligands for it, often requires access to a bound (holo) and an unbound (apo) state of the protein. Resources for the quick and easy retrieval of such conformations are severely limited. Apo-Holo Juxtaposition (AHoJ), is a web application for retrieving apo-holo structure pairs for user-defined ligands. Given a query structure and one or more user-specified ligands, it retrieves all other structures of the same protein that feature the same binding site(s), aligns them, and examines the superimposed binding sites to determine whether each structure is apo or holo, in reference to the query. The resulting superimposed datasets of apo-holo pairs can be visualized and downloaded for further analysis. AHoJ accepts multiple input queries, allowing the creation of customized apo-holo datasets. AVAILABILITY AND IMPLEMENTATION: Freely available for non-commercial use at http://apoholo.cz. Source code available at https://github.com/cusbg/AHoJ-project. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Catastrophic transitions are ubiquitous in the dynamic progression of complex biological systems; that is, a critical transition at which complex systems suddenly shift from one stable state to another occurs. Identifying such a critical point or tipping point is essential for revealing the underlying mechanism of complex biological systems. However, it is difficult to identify the tipping point since few significant differences in the critical state are detected in terms of traditional static measurements. RESULTS: In this study, by exploring the dynamic changes in gene cooperative effects between the before-transition and critical states, we presented a model-free approach, the directed-network rank score (DNRS), to detect the early-warning signal of critical transition in complex biological systems. The proposed method is applicable to both bulk and single-cell RNA-sequencing (scRNA-seq) data. This computational method was validated by the successful identification of the critical or pre-transition state for both simulated and six real datasets, including three scRNA-seq datasets of embryonic development and three tumor datasets. In addition, the functional and pathway enrichment analyses suggested that the corresponding DNRS signaling biomarkers were involved in key biological processes. AVAILABILITY AND IMPLEMENTATION: The source code is freely available at https://github.com/zhongjiayuan/DNRS. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: In recent years, Deep Learning (DL) has been increasingly used in many fields, in particular in image recognition, due to its ability to solve problems where traditional machine learning algorithms fail. However, building an appropriate DL model from scratch, especially in the context of ecological studies, is a difficult task due to the dynamic nature and morphological variability of living organisms, as well as the high cost in terms of time, human resources and skills required to label a large number of training images. To overcome this problem, Transfer Learning (TL) can be used to improve a classifier by transferring information learnt from many domains thanks to a very large training set composed of various images, to another domain with a smaller amount of training data. To compensate the lack of 'easy-to-use' software optimized for ecological studies, we propose the EcoTransLearn R-package, which allows greater automation in the classification of images acquired with various devices (FlowCam, ZooScan, photographs, etc.), thanks to different TL methods pre-trained on the generic ImageNet dataset. AVAILABILITY AND IMPLEMENTATION: EcoTransLearn is an open-source package. It is implemented in R and calls Python scripts for image classification step (using reticulate and tensorflow libraries). The source code, instruction manual and examples can be found at https://github.com/IFREMER-LERBL/EcoTransLearn. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Antimicrobial peptides (AMPs) have the potential to inhibit multiple types of pathogens and to heal infections. Computational strategies can assist in characterizing novel AMPs from proteome or collections of synthetic sequences and discovering their functional abilities toward different microbial targets without intensive labor. RESULTS: Here, we present a deep learning-based method for computer-aided novel AMP discovery that utilizes the transformer neural network architecture with knowledge from natural language processing to extract peptide sequence information. We implemented the method for two AMP-related tasks: the first is to discriminate AMPs from other peptides, and the second task is identifying AMPs functional activities related to seven different targets (gram-negative bacteria, gram-positive bacteria, fungi, viruses, cancer cells, parasites and mammalian cell inhibition), which is a multi-label problem. In addition, asymmetric loss was adopted to resolve the intrinsic imbalance of dataset, particularly for the multi-label scenarios. The evaluation showed that our proposed scheme achieves the best performance for the first task (96.85% balanced accuracy) and has a more unbiased prediction for the second task (79.83% balanced accuracy averaged across all functional activities) when compared with that of strategies without imbalanced learning or deep learning. AVAILABILITY AND IMPLEMENTATION: The source code and data of this study are available at https://github.com/BiOmicsLab/TransImbAMP. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

SUMMARY: The allosteric modulation of peripheral membrane proteins (PMPs) by targeting protein-membrane interactions with drug-like molecules represents a new promising therapeutic strategy for proteins currently considered undruggable. However, the accessibility of protein-membrane interfaces by small molecules has been so far unexplored, possibly due to the complexity of the interface, the limited protein-membrane structural information and the lack of computational workflows to study it. Herein, we present a pipeline for drugging protein-membrane interfaces using the DREAMM (Drugging pRotein mEmbrAne Machine learning Method) web server. DREAMM works in the back end with a fast and robust ensemble machine learning algorithm for identifying protein-membrane interfaces of PMPs. Additionally, DREAMM also identifies binding pockets in the vicinity of the predicted membrane-penetrating amino acids in protein conformational ensembles provided by the user or generated within DREAMM. AVAILABILITY AND IMPLEMENTATION: DREAMM web server is accessible via https://dreamm.ni4os.eu. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: Understanding the functional consequence of genetic variants, especially the non-coding ones, is important but particularly challenging. Genome-wide association studies (GWAS) or quantitative trait locus analyses may be subject to limited statistical power and linkage disequilibrium, and thus are less optimal to pinpoint the causal variants. Moreover, most existing machine-learning approaches, which exploit the functional annotations to interpret and prioritize putative causal variants, cannot accommodate the heterogeneity of personal genetic variations and traits in a population study, targeting a specific disease. RESULTS: By leveraging paired whole-genome sequencing data and epigenetic functional assays in a population study, we propose a multi-modal deep learning framework to predict genome-wide quantitative epigenetic signals by considering both personal genetic variations and traits. The proposed approach can further evaluate the functional consequence of non-coding variants on an individual level by quantifying the allelic difference of predicted epigenetic signals. By applying the approach to the ROSMAP cohort studying Alzheimer's disease (AD), we demonstrate that the proposed approach can accurately predict quantitative genome-wide epigenetic signals and in key genomic regions of AD causal genes, learn canonical motifs reported to regulate gene expression of AD causal genes, improve the partitioning heritability analysis and prioritize putative causal variants in a GWAS risk locus. Finally, we release the proposed deep learning model as a stand-alone Python toolkit and a web server. AVAILABILITY AND IMPLEMENTATION: https://github.com/lichen-lab/DeepPerVar. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.