GitHub Repository Research Articles

Eelbrain, a Python toolkit for time-continuous analysis with temporal response functions.

Even though human experience unfolds continuously in time, it is not strictly linear; instead, it entails cascading processes building hierarchical cognitive structures. For instance, during speech perception, humans transform a continuously varying acoustic signal into phonemes, words, and meaning, and these levels all have distinct but interdependent temporal structures. Time-lagged regression using temporal response functions (TRFs) has recently emerged as a promising tool for disentangling electrophysiological brain responses related to such complex models of perception. Here, we introduce the Eelbrain Python toolkit, which makes this kind of analysis easy and accessible. We demonstrate its use, using continuous speech as a sample paradigm, with a freely available EEG dataset of audiobook listening. A companion GitHub repository provides the complete source code for the analysis, from raw data to group-level statistics. More generally, we advocate a hypothesis-driven approach in which the experimenter specifies a hierarchy of time-continuous representations that are hypothesized to have contributed to brain responses, and uses those as predictor variables for the electrophysiological signal. This is analogous to a multiple regression problem, but with the addition of a time dimension. TRF analysis decomposes the brain signal into distinct responses associated with the different predictor variables by estimating a multivariate TRF (mTRF), quantifying the influence of each predictor on brain responses as a function of time(-lags). This allows asking two questions about the predictor variables: (1) Is there a significant neural representation corresponding to this predictor variable? And if so, (2) what are the temporal characteristics of the neural response associated with it? Thus, different predictor variables can be systematically combined and evaluated to jointly model neural processing at multiple hierarchical levels. We discuss applications of this approach, including the potential for linking algorithmic/representational theories at different cognitive levels to brain responses through computational models with appropriate linking hypotheses.

A dynamic and collaborative database for morphogeometric information of trilobites

Modern morphometric-based approaches provide valuable metrics to quantify and understand macroevolutionary and macroecological patterns and processes. Here we describe TriloMorph, an openly accessible database for morpho-geometric information of trilobites, together with a landmark acquisition protocol. In addition to morphological traits, the database contains contextual data on chronostratigraphic age, geographic location, taxonomic information and lithology of landmarked specimens. In this first version, the dataset has broad taxonomic and temporal coverage and comprises more than 55% of all trilobite genera and 85% of families recorded in the Paleobiology Database through the Devonian. We provide a release of geometric morphometric data of 277 specimens linked to published references. Additionally, we established a Github repository for constant input of morphometric data by multiple contributors and present R functions that help with data retrieval and analysis. This is the first attempt of an online, dynamic and collaborative morphometric repository. By bringing this information into a single open database we enhance the possibility of performing global palaeobiological research, providing a major complement to current occurrence-based databases.

The two-point correlation function covariance with fewer mocks

ABSTRACT We present fitcov an approach for accurate estimation of the covariance of two-point correlation functions that requires fewer mocks than the standard mock-based covariance. This can be achieved by dividing a set of mocks into jackknife regions and fitting the correction term first introduced in Mohammad & Percival (2022), such that the mean of the jackknife covariances corresponds to the one from the mocks. This extends the model beyond the shot-noise limited regime, allowing it to be used for denser samples of galaxies. We test the performance of our fitted jackknife approach, both in terms of accuracy and precision, using lognormal mocks with varying densities and approximate EZmocks mimicking the Dark Energy Spectroscopic Instrument LRG and ELG samples in the redshift range of z = [0.8, 1.1]. We find that the Mohammad–Percival correction produces a bias in the two-point correlation function covariance matrix that grows with number density and that our fitted jackknife approach does not. We also study the effect of the covariance on the uncertainty of cosmological parameters by performing a full-shape analysis. We demonstrate that our fitted jackknife approach based on 25 mocks can recover unbiased and as precise cosmological parameters as the ones obtained from a covariance matrix based on 1000 or 1500 mocks, while the Mohammad–Percival correction produces uncertainties that are twice as large. The number of mocks required to obtain an accurate estimation of the covariance for the two-point correlation function is therefore reduced by a factor of 40–60. The fitcov code that accompanies this paper is available at this GitHub repository.

DeepFPlearn +: enhancing toxicity prediction across the chemical universe using graph neural networks.

Sophisticated approaches for the in silico prediction of toxicity are required to support the risk assessment of chemicals. The number of chemicals on the global chemical market and the speed of chemical innovation stand in massive contrast to the capacity for regularizing chemical use. We recently proved our ready-to-use application deepFPlearn as a suitable approach for this task. Here, we present its extension deepFPlearn+ incorporating (i) a graph neural network to feed our AI with a more sophisticated molecular structure representation and (ii) alternative train-test splitting strategies that involve scaffold structures and the molecular weights of chemicals. We show that the GNNs outperform the previous model substantially and that our models can generalize on unseen data even with a more robust and challenging test set. Therefore, we highly recommend the application of deepFPlearn+ on the chemical inventory to prioritize chemicals for experimental testing or any chemical subset of interest in monitoring studies. The software is compatible with python 3.6 or higher, and the source code can be found on our GitHub repository: https://github.com/yigbt/deepFPlearn. The data underlying this article are available in Zenodo, and can be accessed with the link below: https://zenodo.org/record/8146252. Detailed installation guides via Docker, Singularity, and Conda are provided within the repository for operability across all operating systems.

SilenceREIN: seeking silencers on anchors of chromatin loops by deep graph neural networks.

Silencers are repressive cis-regulatory elements that play crucial roles in transcriptional regulation. Experimental methods for identifying silencers are always costly and time-consuming. Computational methods, which relies on genomic sequence features, have been introduced as alternative approaches. However, silencers do not have significant epigenomic signature. Therefore, we explore a new way to computationally identify silencers, by incorporating chromatin structural information. We propose the SilenceREIN method, which focuses on finding silencers on anchors of chromatin loops. By using graph neural networks, we extracted chromatin structural information from a regulatory element interaction network. SilenceREIN integrated the chromatin structural information with linear genomic signatures to find silencers. The predictive performance of SilenceREIN is comparable or better than other states-of-the-art methods. We performed a genome-wide scanning to systematically find silencers in human genome. Results suggest that silencers are widespread on anchors of chromatin loops. In addition, enrichment analysis of transcription factor binding motif support our prediction results. As far as we can tell, this is the first attempt to incorporate chromatin structural information in finding silencers. All datasets and source codes of SilenceREIN have been deposited in a GitHub repository (https://github.com/JianHPan/SilenceREIN).

Directional intensified feature description using tertiary filtering for augmented reality tracking

Augmented Reality (AR) is applied in almost every field, and a few, but not limited, are engineering, medical, gaming and internet of things. The application of image tracking is inclusive in all these mentioned fields. AR uses image tracking to localize and register the position of the user/AR device for superimposing the virtual image into the real-world. In general terms, tracking the image enhance the users’ experience. However, in the image tracking application, establishing the interface between virtual realm and the physical world has many shortcomings. Many tracking systems are available, but it lacks in robustness and efficiency. The robustness of the tracking algorithm, is the challenging task of implementation. This study aims to enhance the users’ experience in AR by describing an image using Directional Intensified Features with Tertiary Filtering. This way of describing the features improve the robustness, which is desired in the image tracking. A feature descriptor is robust, in the sense that it does not compromise, when the image undergoes various transformations. This article, describes the features based on the Directional Intensification using Tertiary Filtering (DITF). The robustness of the algorithm is improved, because of the inherent design of Tri-ocular, Bi-ocular and Dia-ocular filters that can intensify the features in all required directions. The algorithm’s robustness is verified with respect to various image transformations. The oxford dataset is used for performance analysis and validation. DITF model is designed to achieve the repeatability score of illumination-variation , blur changes and view-point variation, as 100%, 100% and 99% respectively. The comparative analysis has been performed in terms of precision and re-call. DITF outperforms the state-of-the-art descriptors, namely, BEBLID, BOOST, HOG, LBP, BRISK and AKAZE. An Implementation of DITF source code is available in the following GitHub repository: github.com/Johnchristopherclement/Directional-Intensified-Feature-Descriptor.

Photos and rendered images of LEGO bricks

The paper describes a collection of datasets containing both LEGO brick renders and real photos. The datasets contain around 155,000 photos and nearly 1,500,000 renders. The renders aim to simulate real-life photos of LEGO bricks allowing faster creation of extensive datasets. The datasets are publicly available via the Gdansk University of Technology “Most Wiedzy” institutional repository. The source files of all tools used during the creation of the dataset were made publicly available via GitHub repositories. The images, both photos and the renders were annotated with the unique brick ID and category from the official LEGO catalog. The proposed datasets are stored in easy-to-read formats and are labeled via directory structure allowing easy manipulation and conversion of metadata to other formats.

CardioRVAR: A New R Package and Shiny Application for the Evaluation of Closed-Loop Cardiovascular Interactions.

CardioRVAR is a new R package designed for the complete evaluation of closed-loop cardiovascular interactions and baroreflex sensitivity estimated from continuous non-invasive heart rate and blood pressure recordings. In this work, we highlight the importance of this software tool in the context of human cardiovascular and autonomic neurophysiology. A summary of the main algorithms that CardioRVAR uses are reviewed, and the workflow of this package is also discussed. We present the results obtained from this tool after its application in three clinical settings. These results support the potential clinical and scientific applications of this tool. The open-source tool can be downloaded from a public GitHub repository, as well as its specific Shiny application, CardioRVARapp. The open-source nature of the tool may benefit the future continuation of this work.

Exploring the Integration of Machine Learning Models in Programming Languages on GitHub: Impact on Compatibility to Address Them

GitHub repositories are often used for collaborative development, allowing multiple developers to work on the same codebase and contribute their changes. Each repository is typically associated with a specific project, and it can contain everything from code files to documentation, bug reports, and feature requests. Depending on the context, it can contain files, directories, other resources related to a project, and it is often associated with a particular programming language. By default, GitHub automatically detects the primary programming language used in a repository based on the file extensions and content within the repository. However, this detection is not true all the time; there are some potential issues to consider. One of these problems is that the detected language may not accurately reflect the actual programming languages used in the project, especially if the project utilizes multiple programming languages or has undergone language migrations. In this study, we apply an alternative technology to resolve problems with classifying the programming language of a GitHub repository by analysing file extensions and identifying all programming languages used in the project. We also determine the appropriate primary programming language for the repository. This paper investigates how this technology can address the issues surrounding GitHub’s automatic detection of a repository’s primary programming language and how it can provide information on all the programming languages used in a project.

Fake news detection: Taxonomy and comparative study

The proliferation of social networks has presented a significant challenge in combating the pervasive issue of fake news within modern societies. Due to the large amount of information and news produced daily in text, audio, and video, the validation and verification of this information have become crucial tasks. Leveraging advancements in artificial intelligence, distinguishing between fake news and factual information through automatic fake news detection systems has become more feasible. Automatic fake news detection has been explored from diverse perspectives, employing various feature extraction and classification models. Nonetheless, empirical evaluations, categorization, and comparisons of existing techniques for handling this problem remain limited. In this paper, we revisit the definitions and perspectives of fake news and propose an updated taxonomy for the field based on multiple criteria: (1) Type of features used in fake news detection; (2) Fake news detection perspectives; (3) Feature representation methods; and (4) Classification approaches. Moreover, we conduct an extensive empirical study to evaluate several feature representation techniques and classification approaches based on accuracy and computational cost. Our experimental results demonstrate that the optimal feature extraction techniques vary depending on the characteristics of the dataset. Notably, context-dependent models based on transformer models consistently exhibit superior performance. Additionally, employing transformer models as feature extraction methods, rather than solely fine-tuning the network for the downstream task, improves overall performance. Through extensive error analysis, we identify that a combination of feature representation methods and classification algorithms, including classical ones, offer complementary aspects and should be considered for achieving better generalization performance while maintaining a relatively low computational cost. For further details, including source codes, figures, and datasets, please refer to our project’s GitHub repository: [https://github.com/FFarhangian/Fake-news-detection-Comparative-Study].

An effective and robust genetic algorithm with hybrid multi-strategy and mechanism for airport gate allocation

Gate allocation is a combinatorial scheduling problem with multi-constraint and multi-objective. It is challenging to solve this problem when the size of flights increases continuously. As an adaptive technology with a random search ability, the genetic algorithm (GA) has been widely used for resource scheduling and combinatorial optimization. However, it is prone to slow convergence and falling into local optimal solutions. Therefore, an effective and robust GA based on hybrid multi-strategy of reverse learning, interval probability mutation, and phagocytosis mechanism, called RPIP-GA, is proposed to implement a new airport gate-allocation method. In the RPIP-GA, the population is divided into several subpopulations based on the fitness values of all individuals to prevent population degradation and improve population quality. The reverse learning strategy with elite retention is designed to initialize the population, expand the global search space, improve the quality of the original solution, and increase the population diversity. The phagocytosis mechanism is employed to implement a crossover operation to enhance the convergence rate and local search ability. An interval probability mutation technique is designed to improve the local search ability in the early stage and prevent falling into the local optimum in the later stage. The effectiveness of the RPIP-GA is validated using 45 complex functions selected from the benchmark functions and CEC 2017, 22 real-world engineering problems selected from CEC 2011, and an actual gate allocation problem via comparisons with the GA, PSO, GSA, DNLGSA, WMSDE, PADE, HIRCGA, and other algorithms. The experimental results show that the RPIP-GA can obtain optimal results with improved stability in most cases. The maximum allocation rate of actual airport gates reaches 98%, and the average convergence accuracy increases by 19% compared with that of the GA. The data used in the study is publicly available from the GitHub repository (https://github.com/xiaocangjiu).

Re-defining wearable robots: a multidisciplinary approach towards a unified terminology

Effective communication is especially important in the wearable robots (WRs) community, which encloses a great variety of devices across different application domains, e.g., healthcare, occupational, and consumer. In this paper we present a vocabulary of terms with the aim to create a common understanding of terms and concepts among the different fields of expertise relevant in the WRs community. Our goal is to develop shared documentation that could serve as a reference to facilitate the use of accepted definitions in the field. The presented vocabulary is the result of different focus group discussions among experts in the field. The resulting document was then validated by presenting it to the WR community through an online survey. The results of the survey highlight a strong agreement in terms of acceptance of the vocabulary, its usefulness, and applicability of the proposed definitions as well as an overall appreciation for its purpose and target. This work represents a pilot study providing unique material for the WR community, encouraging the use of shared agreed definitions. The reported version of the vocabulary has been made available as a live document in a github repository, for public commenting and further improvements.

Uncertainty aware neural network from similarity and sensitivity

Recent uncertainty quantification approaches lack transparency. Algorithms often perform poorly in an input domain and the reason for poor performance remains unknown. Therefore, we present a neural network training method that considers similar samples with sensitivity awareness in this paper. In the proposed NN training method for UQ, first, we train a shallow NN for the point prediction. Then, we train another NN to predict absolute differences between targets and predictions. In the next step, we select each sample in the training set one by one and compute both prediction and error sensitivities. Then we select similar samples with sensitivity consideration and save indexes of similar samples. The ranges of an input parameter become narrower when the output is highly sensitive to that parameter. After that, we construct initial uncertainty bounds (UB) by considering the distribution of sensitivity-aware similar samples. Prediction intervals (PIs) from initial uncertainty bounds are larger and cover more samples than required. Therefore, we train bound correction NN. As following all the steps for finding UB for each sample requires a lot of computation and memory access, we train a UB computation NN. The UB computation NN takes an input sample and provides an uncertainty bound. The UB computation NN is the final product of the proposed approach. We have achieved superior performance in most situations through the proposed transparent method. Scripts of the proposed method are available in the following GitHub repository: https://github.com/dipuk0506/UQ

On moments of truncated multivariate normal/independent distributions

Multivariate normal/independent (MNI) distributions contain many renowned heavy-tailed distributions such as the multivariate t, multivariate slash, multivariate contaminated normal, multivariate variance-gamma, and multivariate double exponential distributions. A frequent problem encountered in statistical analysis is the occurrence of truncated observations and non-normality such that theoretical moments are required for the estimation of the truncated multivariate normal/independent (TMNI) distributions. This paper is dedicated to deriving explicit expressions for the moments of the TMNI distributions with supports confined within a hyper-rectangle. A Monte Carlo experiment is undertaken to validate to the correctness of the proposed formulae for five selected members of the TMNI distributions. R scripts and data to reproduce the results are available in the GitHub repository.

A community convention for ecological forecasting: Output files and metadata version 1.0

AbstractThis paper summarizes the open community conventions developed by the Ecological Forecasting Initiative (EFI) for the common formatting and archiving of ecological forecasts and the metadata associated with these forecasts. Such open standards are intended to promote interoperability and facilitate forecast communication, distribution, validation, and synthesis. For output files, we first describe the convention conceptually in terms of global attributes, forecast dimensions, forecasted variables, and ancillary indicator variables. We then illustrate the application of this convention to the two file formats that are currently preferred by the EFI, netCDF (network common data form), and comma‐separated values (CSV), but note that the convention is extensible to future formats. For metadata, EFI's convention identifies a subset of conventional metadata variables that are required (e.g., temporal resolution and output variables) but focuses on developing a framework for storing information about forecast uncertainty propagation, data assimilation, and model complexity, which aims to facilitate cross‐forecast synthesis. The initial application of this convention expands upon the Ecological Metadata Language (EML), a commonly used metadata standard in ecology. To facilitate community adoption, we also provide a Github repository containing a metadata validator tool and several vignettes in R and Python on how to both write and read in the EFI standard. Lastly, we provide guidance on forecast archiving, making an important distinction between short‐term dissemination and long‐term forecast archiving, while also touching on the archiving of code and workflows. Overall, the EFI convention is a living document that can continue to evolve over time through an open community process.

LCK-SafeScreen-Model: An Advanced Ensemble Machine Learning Approach for Estimating the Binding Affinity between Compounds and LCK Target.

The lymphocyte-specific protein tyrosine kinase (LCK) is a critical target in leukemia treatment. However, potential off-target interactions involving LCK can lead to unintended consequences. This underscores the importance of accurately predicting the inhibitory reactions of drug molecules with LCK during the research and development stage. To address this, we introduce an advanced ensemble machine learning technique designed to estimate the binding affinity between molecules and LCK. This comprehensive method includes the generation and selection of molecular fingerprints, the design of the machine learning model, hyperparameter tuning, and a model ensemble. Through rigorous optimization, the predictive capabilities of our model have been significantly enhanced, raising test R2 values from 0.644 to 0.730 and reducing test RMSE values from 0.841 to 0.732. Utilizing these advancements, our refined ensemble model was employed to screen an MCE -like drug library. Through screening, we selected the top ten scoring compounds, and tested them using the ADP-Glo bioactivity assay. Subsequently, we employed molecular docking techniques to further validate the binding mode analysis of these compounds with LCK. The exceptional predictive accuracy of our model in identifying LCK inhibitors not only emphasizes its effectiveness in projecting LCK-related safety panel predictions but also in discovering new LCK inhibitors. For added user convenience, we have also established a webserver, and a GitHub repository to share the project.

The utilisation of open research data repositories for storing and sharing research data in higher learning institutions in Tanzania

PurposeThe study aims to investigate the utilisation of open research data repositories (RDRs) for storing and sharing research data in higher learning institutions (HLIs) in Tanzania.Design/methodology/approachA survey research design was employed to collect data from postgraduate students at the Nelson Mandela African Institution of Science and Technology (NM-AIST) in Arusha, Tanzania. The data were collected and analysed quantitatively and qualitatively. A census sampling technique was employed to select the sample size for this study. The quantitative data were analysed using the Statistical Package for the Social Sciences (SPSS), whilst the qualitative data were analysed thematically.FindingsLess than half of the respondents were aware of and were using open RDRs, including Zenodo, DataVerse, Dryad, OMERO, GitHub and Mendeley data repositories. More than half of the respondents were not willing to share research data and cited a lack of ownership after storing their research data in most of the open RDRs and data security. HILs need to conduct training on using trusted repositories and motivate postgraduate students to utilise open repositories (ORs). The challenges for underutilisation of open RDRs were a lack of policies governing the storage and sharing of research data and grant constraints.Originality/valueResearch data storage and sharing are of great interest to researchers in HILs to inform them to implement open RDRs to support these researchers. Open RDRs increase visibility within HILs and reduce research data loss, and research works will be cited and used publicly. This paper identifies the potential for additional studies focussed on this area.

DsRID: in silico identification of dsRNA regions using long-read RNA-seq data.

Double-stranded RNAs (dsRNAs) are potent triggers of innate immune responses upon recognition by cytosolic dsRNA sensor proteins. Identification of endogenous dsRNAs helps to better understand the dsRNAome and its relevance to innate immunity related to human diseases. Here, we report dsRID (double-stranded RNA identifier), a machine-learning-based method to predict dsRNA regions in silico, leveraging the power of long-read RNA-sequencing (RNA-seq) and molecular traits of dsRNAs. Using models trained with PacBio long-read RNA-seq data derived from Alzheimer's disease (AD) brain, we show that our approach is highly accurate in predicting dsRNA regions in multiple datasets. Applied to an AD cohort sequenced by the ENCODE consortium, we characterize the global dsRNA profile with potentially distinct expression patterns between AD and controls. Together, we show that dsRID provides an effective approach to capture global dsRNA profiles using long-read RNA-seq data. Software implementation of dsRID, and genomic coordinates of regions predicted by dsRID in all samples are available at the GitHub repository: https://github.com/gxiaolab/dsRID.

Genome-scale metabolic models consistently predict in vitro characteristics of Corynebacterium striatum.

Introduction: Genome-scale metabolic models (GEMs) are organism-specific knowledge bases which can be used to unravel pathogenicity or improve production of specific metabolites in biotechnology applications. However, the validity of predictions for bacterial proliferation in in vitro settings is hardly investigated. Methods: The present work combines in silico and in vitro approaches to create and curate strain-specific genome-scale metabolic models of Corynebacterium striatum. Results: We introduce five newly created strain-specific genome-scale metabolic models (GEMs) of high quality, satisfying all contemporary standards and requirements. All these models have been benchmarked using the community standard test suite Metabolic Model Testing (MEMOTE) and were validated by laboratory experiments. For the curation of those models, the software infrastructure refineGEMs was developed to work on these models in parallel and to comply with the quality standards for GEMs. The model predictions were confirmed by experimental data and a new comparison metric based on the doubling time was developed to quantify bacterial growth. Discussion: Future modeling projects can rely on the proposed software, which is independent of specific environmental conditions. The validation approach based on the growth rate calculation is now accessible and closely aligned with biological questions. The curated models are freely available via BioModels and a GitHub repository and can be used. The open-source software refineGEMs is available from https://github.com/draeger-lab/refinegems.

MosaiCatcher v2: a single-cell structural variations detection and analysis reference framework based on Strand-seq.

Single-cell DNA template strand sequencing (Strand-seq) allows a range of various genomic analysis including chromosome length haplotype phasing and structural variation (SV) calling in individual cells. Here, we present MosaiCatcher v2, a standardized workflow and reference framework for single-cell SV detection using Strand-seq. This framework introduces a range of functionalities, including: an automated upstream Quality Control (QC) and assembly sub-workflow that relies on multiple genome assemblies and incorporates a multistep normalization module, integration of the single-cell nucleosome occupancy and genetic variation analysis SV functional characterization and of the ArbiGent SV genotyping modules, platform portability, as well as a user-friendly and shareable web report. These new features of MosaiCatcher v2 enable reproducible computational processing of Strand-seq data, which are increasingly used in human genetics and single-cell genomics, toward production environments. MosaiCatcher v2 is compatible with both container and conda environments, ensuring reproducibility and robustness and positioning the framework as a cornerstone in computational processing of Strand-seq data. MosaiCatcher v2 is a standardized workflow, implemented using the Snakemake workflow management system. The pipeline is available on GitHub: https://github.com/friendsofstrandseq/mosaicatcher-pipeline/ and on the snakemake-workflow-catalog: https://snakemake.github.io/snakemake-workflow-catalog/?usage=friendsofstrandseq/mosaicatcher-pipeline. Strand-seq example input data used in the publication can be found in the Data availability statement. Additionally, a lightweight dataset for test purposes can be found on the GitHub repository.