The precise prediction of transcription factor binding sites (TFBSs) is pivotal for unraveling the gene regulatory networks underlying biological processes. While numerous tools have emerged for in silico TFBS prediction in recent years, the evolving landscape of computational biology necessitates thorough assessments of tool performance to ensure accuracy and reliability. Only a limited number of studies have been conducted to evaluate the performance of TFBS prediction tools comprehensively. Thus, the present study focused on assessing twelve widely used TFBS prediction tools and four de novo motif discovery tools using a benchmark dataset comprising real, generic, Markov, and negative sequences. TFBSs of Arabidopsis thaliana and Homo sapiens genomes downloaded from the JASPAR database were implanted in these sequences and the performance of tools was evaluated using several statistical parameters at different overlap percentages between the lengths of known and predicted binding sites.

Insertions and deletions (indels) play a significant role in genome evolution across species. Realistic modelling of indel evolution is challenging and is still an open research question. Several attempts have been made to explicitly model multi-character (long) indels, such as TKF92, by relaxing the site independence assumption and introducing fragments. However, these methods are computationally expensive. On the other hand, the Poisson Indel Process (PIP) assumes site independence but allows one to infer single-character indels on the phylogenetic tree, distinguishing insertions from deletions. PIP's marginal likelihood computation has linear time complexity, enabling ancestral sequence reconstruction (ASR) with indels in linear time. Recently, we developed ARPIP, an ASR method using PIP, capable of inferring indel events with explicit evolutionary interpretations. Here, we investigate the effect of the single-character indel assumption on reconstructed ancestral sequences on mammalian protein orthologs and on simulated data. We show that ARPIP's ancestral estimates preserve the gap length distribution observed in the input alignment. In mammalian proteins the lengths of inserted segments appear to be substantially longer compared to deleted segments. Further, we confirm the well-established deletion bias observed in real data. To date, ARPIP is the only ancestral reconstruction method that explicitly models insertion and deletion events over time. Given a good quality input alignment, it can capture ancestral long indel events on the phylogeny.

Rare copy number variants (CNVs) significantly influence the human genome and may contribute to disease susceptibility. High-throughput SNP genotyping platforms provide data that can be used for CNV detection, but it requires the complex pipelining of bioinformatic tools. Here, we propose a flexible bioinformatic pipeline for rare CNV analysis from human SNP array data.

Graphical representations are useful to model complex data in general and biological interactions in particular. Our main motivation is the comparison of metabolic networks in the wider context of developing noninvasive accurate diagnostic tools. However, comparison and classification of graphs is still extremely challenging, although a number of highly efficient methods such as graph neural networks were developed in the recent decade. Important aspects are still lacking in graph classification: interpretability and guarantees on classification quality, i.e., control of the risk level or false discovery rate control.

The rewiring of molecular interactions in various conditions leads to distinct phenotypic outcomes. Differential network analysis (DINA) is dedicated to exploring these rewirings within gene and protein networks. Leveraging statistical learning and graph theory, DINA algorithms scrutinize alterations in interaction patterns derived from experimental data.

Identification of drug-target interactions is an indispensable part of drug discovery. While conventional shallow machine learning and recent deep learning methods based on chemogenomic properties of drugs and target proteins have pushed this prediction performance improvement to a new level, these methods are still difficult to adapt to novel structures. Alternatively, large-scale biological and pharmacological data provide new ways to accelerate drug-target interaction prediction.

Bioactive peptides are important bioactive molecules composed of short-chain amino acids that play various crucial roles in the body, such as regulating physiological processes and promoting immune responses and antibacterial effects. Due to their significance, bioactive peptides have broad application potential in drug development, food science, and biotechnology. Among them, understanding their biological mechanisms will contribute to new ideas for drug discovery and disease treatment.

In unforeseen situations, such as nuclear power plant's or civilian radiation accidents, there is a need for effective and computationally inexpensive methods to determine the expression level of a selected gene panel, allowing for rough dose estimates in thousands of donors. The new generation in-situ mapper, fast and of low energy consumption, working at the level of single nanopore output, is in demand. We aim to create a sequence identification tool that utilizes natural language processing techniques and ensures a high level of negative predictive value (NPV) compared to the classical approach.

Over the last decade the drop in short-read sequencing costs has allowed experimental techniques utilizing sequencing to address specific biological questions to proliferate, oftentimes outpacing standardized or effective analysis approaches for the data generated. There are growing amounts of bacterial 3'-end sequencing data, yet there is currently no commonly accepted analysis methodology for this datatype. Most data analysis approaches are somewhat ad hoc and, despite the presence of substantial signal within annotated genes, focus on genomic regions outside the annotated genes (e.g. 3' or 5' UTRs). Furthermore, the lack of consistent systematic analysis approaches, as well as the absence of genome-wide ground truth data, make it impossible to compare conclusions generated by different labs, using different organisms.

The preservation of soil health is a critical challenge in the 21st century due to its significant impact on agriculture, human health, and biodiversity. We provide one of the first comprehensive investigations into the predictive potential of machine learning models for understanding the connections between soil and biological phenotypes. We investigate an integrative framework performing accurate machine learning-based prediction of plant performance from biological, chemical, and physical properties of the soil via two models: random forest and Bayesian neural network.

Plasmids play a major role in the transfer of antimicrobial resistance (AMR) genes among bacteria via horizontal gene transfer. The identification of plasmids in short-read assemblies is a challenging problem and a very active research area. Plasmid binning aims at detecting, in a draft genome assembly, groups (bins) of contigs likely to originate from the same plasmid. Several methods for plasmid binning have been developed recently, such as PlasBin-flow, HyAsP, gplas, MOB-suite, and plasmidSPAdes. This motivates the problem of evaluating the performances of plasmid binning methods, either against a given ground truth or between them.

Genomic sequence similarity comparison is a crucial research area in bioinformatics. Multiple Sequence Alignment (MSA) is the basic technique used to identify regions of similarity between sequences, although MSA tools are widely used and highly accurate, they are often limited by computational complexity, and inaccuracies when handling highly divergent sequences, which leads to the development of alignment-free (AF) algorithms.

Antimicrobial peptides (AMPs) are a promising class of antimicrobial drugs due to their broad-spectrum activity against microorganisms. However, their clinical application is limited by their potential to cause hemolysis, the destruction of red blood cells. To address this issue, we propose a deep learning model based on convolutional neural networks (CNNs) for predicting the hemolytic activity of AMPs. Peptide sequences are represented using one-hot encoding, and the CNN architecture consists of multiple convolutional and fully connected layers. The model was trained on six different datasets: HemoPI-1, HemoPI-2, HemoPI-3, RNN-Hem, Hlppredfuse, and AMP-Combined, achieving Matthew's correlation coefficients of 0.9274, 0.5614, 0.6051, 0.6142, 0.8799, and 0.7484, respectively. Our model outperforms previously reported methods and can facilitate the development of novel AMPs with reduced hemolytic activity, which is crucial for their therapeutic use in treating bacterial infections.

Jointly analyzing multiple phenotype/traits may increase power in genetic association studies by aggregating weak genetic effects. The chance that at least one phenotype is missing increases exponentially as the number of phenotype increases especially for a real dataset. It is a common practice to discard individuals with missing phenotype or phenotype with a large proportion of missing values. Such a discarding method may lead to a loss of power or even an insufficient sample size for analysis. To our knowledge, many existing phenotype imputing methods are built on multivariate normal assumptions for analysis. Violation of these assumptions may lead to inflated type I errors or even loss of power in some cases. To overcome these limitations, we propose a novel phenotype imputation method based on a new Gaussian copula model with three different loss functions to address the issue of missing phenotype.

Recently, there has been a growing interest in combining causal inference with machine learning algorithms. Double machine learning model (DML), as an implementation of this combination, has received widespread attention for their expertise in estimating causal effects within high-dimensional complex data. However, the DML model is sensitive to the presence of outliers and heavy-tailed noise in the outcome variable. In this paper, we propose the robust double machine learning (RDML) model to achieve a robust estimation of causal effects when the distribution of the outcome is contaminated by outliers or exhibits symmetrically heavy-tailed characteristics.

Networks have emerged as a natural data structure to represent relations among entities. Proteins interact to carry out cellular functions and protein-Protein interaction network analysis has been employed for understanding the cellular machinery. Advances in genomics technologies enabled the collection of large data that annotate proteins in interaction networks. Integrative analysis of interaction networks with gene expression and annotations enables the discovery of context-specific complexes and improves the identification of functional modules and pathways. Extracting subnetworks whose vertices are connected and have high attribute similarity have applications in diverse domains. We present an enumeration approach for mining sets of connected and cohesive subgraphs, where vertices in the subgraphs have similar attribute profile. Due to the large number of cohesive connected subgraphs and to overcome the overlap among these subgraphs, we propose an algorithm for enumerating a set of representative subgraphs, the set of all closed subgraphs. We propose pruning strategies for efficiently enumerating the search tree without missing any pattern or reporting duplicate subgraphs. On a real protein-protein interaction network with attributes representing the dysregulation profile of genes in multiple cancers, we mine closed cohesive connected subnetworks and show their biological significance. Moreover, we conduct a runtime comparison with existing algorithms to show the efficiency of our proposed algorithm.

The integration of multi-omics data through deep learning has greatly improved cancer subtype classification, particularly in feature learning and multi-omics data integration. However, key challenges remain in embedding sample structure information into the feature space and designing flexible integration strategies.

The explosive growth of next-generation sequencing data has resulted in ultra-large-scale datasets and significant computational challenges. As the cost of next-generation sequencing (NGS) has decreased, the amount of genomic data has surged globally. However, the cost and complexity of the computational resources required continue to be substantial barriers to leveraging big data. A promising solution to these computational challenges is cloud computing, which provides researchers with the necessary CPUs, memory, storage, and software tools.

Advances in transcriptional profiling methods have enabled the discovery of molecular subtypes within and across traditional tissue-based cancer classifications. Such molecular subgroups hold potential for improving patient outcomes by guiding treatment decisions and revealing physiological distinctions and targetable pathways. Computational methods for stratifying transcriptomic data into molecular subgroups are increasingly abundant. However, assigning samples to these subtypes and other transcriptionally inferred predictions is time-consuming and requires significant bioinformatics expertise. To address this need, we recently reported "ClassifieR," a flexible, interactive cloud application for the functional annotation of colorectal and breast cancer transcriptomes. Here, we report "ClassifieR 2.0" which introduces additional modules for the molecular subtyping of prostate and high-grade serous ovarian cancer (HGSOC).

Viral proteins that evade the host's innate immune response play a crucial role in pathogenesis, significantly impacting viral infections and potential therapeutic strategies. Identifying these proteins through traditional methods is challenging and time-consuming due to the complexity of virus-host interactions. Leveraging advancements in computational biology, we present VirusHound-II, a novel tool that utilizes machine learning techniques to predict viral proteins evading the innate immune response with high accuracy. We evaluated a comprehensive range of machine learning models, including ensemble methods, neural networks, and support vector machines. Using a dataset of 1337 viral proteins known to evade the innate immune response (VPEINRs) and an equal number of non-VPEINRs, we employed pseudo amino acid composition as the molecular descriptor. Our methodology involved a tenfold cross-validation strategy on 80% of the data for training, followed by testing on an independent dataset comprising the remaining 20%. The random forest model demonstrated superior performance metrics, achieving 0.9290 accuracy, 0.9283 F1 score, 0.9354 precision, and 0.9213 sensitivity in the independent testing phase. These results establish VirusHound-II as an advancement in computational virology, accessible via a user-friendly web application. We anticipate that VirusHound-II will be a crucial resource for researchers, enabling the rapid and reliable prediction of viral proteins evading the innate immune response. This tool has the potential to accelerate the identification of therapeutic targets and enhance our understanding of viral evasion mechanisms, contributing to the development of more effective antiviral strategies and advancing our knowledge of virus-host interactions.

RNA 5-methyluridine (m5U) modifications play a crucial role in biological processes, making their accurate identification a key focus in computational biology. This paper introduces Deep-m5U, a robust predictor designed to enhance the prediction of m5U modifications. The proposed method, named Deep-m5U, utilizes a hybrid pseudo-K-tuple nucleotide composition (PseKNC) for sequence formulation, a Shapley Additive exPlanations (SHAP) algorithm for discriminant feature selection, and a deep neural network (DNN) as the classifier.