MOTIVATION: Molecular dynamics (MD) simulations provide detailed atomistic insights into biomolecular processes, but comparing independent trajectories remains challenging due to stochastic divergence. Misaligned simulations can obscure shared mechanisms or exaggerate differences, limiting reproducibility and mechanistic interpretation. A generalizable, unsupervised method for synchronizing and comparing MD trajectories across systems and conditions is, therefore, needed. RESULTS: We introduce NetMD, a computational framework that synchronizes and analyzes MD trajectories by integrating graph-based representations with dynamic time warping. Trajectory frames are converted into residue-contact graphs, entropy-filtered to retain variable interactions, and embedded as low-dimensional vectors. NetMD aligns these vectorized trajectories through time-warping barycenter averaging, generating a consensus trajectory while pruning outlier simulations. Applied to transporters, demethylases, and large protein complexes relevant to neurological disease pathways and cancer, NetMD revealed shared multiphase dynamics and identified mutation- or ligand-specific deviations. This unsupervised, time-resolved approach enables direct comparison of MD ensembles across heterogeneous conditions. NetMD is robust and broadly applicable, providing a tool for uncovering conserved patterns and critical divergences in biomolecular dynamics. AVAILABILITY AND IMPLEMENTATION: NetMD is freely available at https://github.com/mazzalab/NetMD.

Recently, the availability of many omics data source has given the rise of modelling biological networks for each individual or patient. Such networks are able to represent individual-specific characteristics, providing insights into the condition of each person. Given a set of networks of individuals, a network representing a particular condition (e.g., an individual with a specific disease) may be seen as an anomaly network. Consequently, the use of Graph Anomaly Detection techniques may support such analysis. Among the others, Generative Adversarial Networks present optimal performances in anomaly detection. This paper presents ADIN (Anomaly Detection in Individual Networks), a framework based on Generative Adversarial Attributed Networks (GAANs) for anomaly detection in convergence/divergence patients attributed networks. Preliminary results on networks generated from computational biology gene expression data demonstrate the effectiveness of our approach in detecting and explaining bladder cancer patients.

Modelling infectious disease spreading is crucial for planning effective containment measures, as shown in the COVID-19 pandemic. The effectiveness of planned measures can also be measured regarding saved lives and economic resources. Therefore, introducing methods able to model the evolution and the impact of measures, as well as planning tailored and updated measures, is a crucial step. Existing models for spreading modelling belong to two main classes: (i) compartmental models based on ordinary differential equations and (ii) contact-based models based on a contact structure using an underlining layer to simulate diffusion. Nevertheless, none of these methods can leverage the high computational power of artificial intelligence and deep learning. We propose a novel framework for simulating and analysing disease progression for these methods. The framework is based on the multiscale simulation of the spreading based on using a multiscale contact model built on top of a diffusion model customised by the user. The evolution of the spreading, modelled as a graph with attributed nodes, is then mapped into a latent space through graph embedding. Finally, deep learning models are used in the latent space to analyse and forecast methods without running expensive computational simulations of the contact-based model.

MOTIVATION: The control of the diffusion of diseases is a critical subject of a broad research area, which involves both clinical and political aspects. It makes wide use of computational tools, such as ordinary differential equations, stochastic simulation frameworks and graph theory, and interaction data, from molecular to social granularity levels, to model the ways diseases arise and spread. The coronavirus disease 2019 (COVID-19) is a perfect testbench example to show how these models may help avoid severe lockdown by suggesting, for instance, the best strategies of vaccine prioritization. RESULTS: Here, we focus on and discuss some graph-based epidemiological models and show how their use may significantly improve the disease spreading control. We offer some examples related to the recent COVID-19 pandemic and discuss how to generalize them to other diseases.

MOTIVATION: Tumor mutational burden (TMB) has been proposed as a predictive biomarker for immunotherapy response in cancer patients, as it is thought to enrich for tumors with high neoantigen load. TMB assessed by whole-exome sequencing is considered the gold standard but remains confined to research settings. In the clinical setting, targeted gene panels sampling various genomic sizes along with diverse strategies to estimate TMB were proposed and no real standard has emerged yet. RESULTS: We provide the community with TMBleR, a tool to measure the clinical impact of various strategies of panel-based TMB measurement. AVAILABILITY AND IMPLEMENTATION: R package and docker container (GPL-3 Open Source license): https://acc-bioinfo.github.io/TMBleR/. Graphical-user interface website: https://bioserver.ieo.it/shiny/app/tmbler. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Hundreds of human proteins were found to establish transient interactions with rather degenerated consensus DNA sequences or motifs. Identifying these motifs and the genomic sites where interactions occur represent one of the most challenging research goals in modern molecular biology and bioinformatics. The last twenty years witnessed an explosion of computational tools designed to perform this task, whose performance has been last compared fifteen years ago. Here, we survey sixteen of them, benchmark their ability to identify known motifs nested in twenty-nine simulated sequence datasets, and finally report their strengths, weaknesses, and complementarity.

Molecular dynamics (MD) simulation allows one to predict the time evolution of a system of interacting particles. It is widely used in physics, chemistry and biology to address specific questions about the structural properties and dynamical mechanisms of model systems. MD earned a great success in genome research, as it proved to be beneficial in sorting pathogenic from neutral genomic mutations. Considering their computational requirements, simulations are commonly performed on HPC computing devices, which are generally expensive and hard to administer. However, variables like the software tool used for modeling and simulation or the size of the molecule under investigation might make one hardware type or configuration more advantageous than another or even make the commodity hardware definitely suitable for MD studies. This work aims to shed lights on this aspect.

24,189 are all the possible non-synonymous amino acid changes potentially affecting the human mitochondrial DNA. Only a tiny subset was functionally evaluated with certainty so far, while the pathogenicity of the vast majority was only assessed in-silico by software predictors. Since these tools proved to be rather incongruent, we have designed and implemented APOGEE, a machine-learning algorithm that outperforms all existing prediction methods in estimating the harmfulness of mitochondrial non-synonymous genome variations. We provide a detailed description of the underlying algorithm, of the selected and manually curated training and test sets of variants, as well as of its classification ability.

Pancreatic cancer is a highly aggressive and chemotherapy-resistant malignant neoplasm. In basal condition, it is characterized by elevated autophagy activity, which is required for tumor growth and that correlates with treatment failure. We analyzed the expression of autophagy related genes in different pancreatic cancer cell lines. A correlation-based network analysis evidenced the sociality and topological roles of the autophagy-related genes after serum starvation. Structural and functional tests identified a core set of autophagy related genes, suggesting different scenarios of autophagic responses to starvation, which may be responsible for the clinical variations associated with pancreatic cancer pathogenesis.

A remarkable degree of genetic variation has been found in the protein-encoding regions of DNA through deep sequencing of samples obtained from thousands of subjects from several populations. Approximately half of the 20 000 single nucleotide polymorphisms present, even in normal healthy subjects, are nonsynonymous amino acid substitutions that could potentially affect protein function. The greatest challenges currently facing investigators are data interpretation and the development of strategies to identify the few gene-coding variants that actually cause or confer susceptibility to disease. A confusing array of options is available to address this problem. Unfortunately, the overall accuracy of these tools at ultraconserved positions is low, and predictions generated by current computational tools may mislead researchers involved in downstream experimental and clinical studies. First, we have presented an updated review of these tools and their primary functionalities, focusing on those that are naturally prone to analyze massive variant sets, to infer some interesting similarities among their results. Additionally, we have evaluated the prediction congruency for real whole-exome sequencing data in a proof-of-concept study on some of these web-based tools.

Next generation sequencers have greatly improved our ability to mine polymorphisms and mutations out of entire (or portions of) genomes. The reliability of their outputs, though, showed to be very related to the sequencing chemistry and to deeply affect the quality of the downstream analyses. We focus here on the two-base color code chemistry of AB SOLiD sequencers and propose a comprehensive quality control methodological and software pipeline. We used existing and custom tools to detect and purge short-reads of some common flaws due to sequencing errors and chemical hitches. We apply them to a cohort of SOLiD 4 runs and measure their joint efficacy in terms of the resulting ability to detect the greatest possible number of true variants.

SUMMARY: The Atlas of UTR Regulatory Activity (AURA) is a manually curated and comprehensive catalog of human mRNA untranslated regions (UTRs) and UTR regulatory annotations. Through its intuitive web interface, it provides full access to a wealth of information on UTRs that integrates phylogenetic conservation, RNA sequence and structure data, single nucleotide variation, gene expression and gene functional descriptions from literature and specialized databases. AVAILABILITY: http://aura.science.unitn.it CONTACT: [email protected]; [email protected] SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

This paper includes a selection of papers presented at the Second International Workshop on High Performance Computational Systems Biology (HiBi 2010), which was colocated with the joint ICGT/SPIN conference, that was held at the University of Twente, Enschede, The Netherlands, on 27-29 September 2010. The selected papers cover a broad range of bioinformatics topics, including medical imaging, Markov clustering, simulation, model checking, and gene networks.

Important achievements in traditional biology has deepened the knowledge about living systems leading to an extensive identification of parts-list of the cell as well as of the interactions among biochemical species responsible for cell's regulation. Such an expanding knowledge also introduces new issues. For example the increasing comprehension of the inter- dependencies between pathways (pathways cross-talk) has resulted, on one hand, in the growth of informational complexity, on the other, in a strong lack of information coherence. The overall grand challenge remains unchanged: to be able to assemble the knowledge of every 'piece' of a system in order to figure out the behavior of the whole (integrative approach). In light of these considerations high performance computing plays a fundamental role in the context of in-silico biology. Stochastic simulation is a renowned analysis tool, which, although widely used, is subject to stringent computational requirements, in particular when dealing with heterogeneous and high dimensional systems. Here we introduce and discuss a methodology aimed at alleviating the burden of simulating complex biological networks. Such a method, which springs from graph theory, is based on the principle of fragmenting the computational space of a simulation trace and delegating the computation of fragments to a number of parallel processes.

Over the years, the long-standing questions concerning disease evolution, plausible cellular behavior and the hidden properties of organisms have become more and more pressing. They sparked a movement of experts from an array of scientific areas and stimulated them to work cooperatively with the ambitious aim of deciphering the language of nature. From the onset, the field of genomics has been built on many important discoveries beginning with the DNA double helix structure in the 1950s, reverse transcriptase in the late 1960s, recombinant DNA and restriction enzymes in the 1970s, and finally, the polymerase chain reaction discovery in the early 1980s. For some time the polymerase chain reaction was particularly revolutionary because it was the only method used to determine the base sequence of DNA. Wishing to understand how key components ‘converse’ in time, space and across multiple organizational levels became one of the driving factors that led scientists to sequence large eukaryotic genomes, such as that of human, and to the discovery of more than 20 000 different genes and of nearly a million proteins within the cell. The transition years between the 1990s and first decade of the 2000s were the time of the collection of such results that were later mathematically formalized and summarized in artificial wired diagrams. Soon after, the idea that the full understanding could be possible by a holistic rather than by an atomistic approach became slowly appealing. Investigation on the assembly of the cell’s elementary parts to form complex structures represented the main activity of a promising new research field named ‘Systems Biology’. Progress in this area required interdisciplinary breakthroughs and close links between wet and dry experimentations. Due to an enormous eager interest, synthetic formalizations (or models) grew rapidly in number and size, models that later could be stored in resource databases easily accessible to anyone of interest. ‘BioModels.Net’ is one of such data resources. It is a database of models ‘curated’ by Le Novère and co-workers. In their paper, they offer an overview of some peculiar characteristics of the data bank in light of the computerized web services they provide to ease the user impact. In spite of this, models became unmanageable by manual inspection because of their size and compelled biologists to reduce drastically the kind and quality of the achievable analysis procedures. Those whose job was that of modeling living systems very soon acknowledged how badly the effectiveness of such procedures is undermined by the complexity of the resulting models, which, in fact, are a reflection of the complexity of nature itself. If computer-driven experimentation had not been introduced almost alongside, any outcome would have been incomplete. Many automatic analysis procedures were made available to the scientific community, such as steady state, bifurcation, robustness analyses, model checking, simulation, etc. With regard to this, ‘CaliBayes’ and ‘BASIS’ represent the efforts made by Wilkinson and co-authors to collect and integrate some tools for calibration, simulation and storage of biological simulation models. Alongside them, Uhrmacher’s group has also proposed a valuable modular framework, ‘James II’. It works to ease the process of designing wetlab experiments through advanced routines and algorithms. Both the goals match: they aim at providing new insights for new models and to feed back the so-called ‘hypothesis-driven science cycle’. Dry laboratories proliferated rapidly. To better describe their biological dynamics of interest from precise perspectives, each defined a new appropriate language. As a first result, this gave rise to an explosion of new software tools along with an overall ‘Babel of voices’, since each one of them talked a different dialect. The need for a common language was becoming a ‘must’. Standard languages (like SBML and CellML) were created with the primary aim of making cooperation among software tools possible. They were able to bridge the gap between wet and dry scientists by broadening the range of the available computational methods. They definitely reached their goal. In the case of modeling, successive developments focused on how to deal with more and more complex systems. Hence, increasing evidence in the constant interplay between components of different biological systems (cross-talk) proved that isolated models are quite unrealistic and that any analysis result is imperfect. Therefore, sets of artificial models of higher chemical density and complexity were merged and analyzed, however, without any significant results. In fact, the exponential growth of the computational power required to deal with huge ‘state-spaces’. This had the simple effect of hindering any of the aforementioned methodologies, and thus causing them to ultimately flop. For this reason Safranek, Brim and Barnat have developed ‘DIVinE’, a model checker made up of a collection of tools designed to exploit the capabilities of the new hardware architectures, as well as to restrain the state space of complex models from exploding. One of the main limitations in managing biological models comes from the fundamental difference between evident high parallelism in biochemical reactions and sequential environments employed for the analysis of these reactions. Such limitations affect all varieties of continuous, deterministic, discrete and stochastic models by undermining the applicability of simulation techniques and the analysis of biological models. Parallel and distributed computing may be deployed to compensate for this lack. They rely on both the intrinsic parallelism of nature and the power of multiprocessing architectures. Indeed, in real life any biological aspect is (to some extent) parallel. If a chemical transformation occurs, it does not take place for two molecules, but, as a principle, for all molecules. In a less gross view, hundreds of independent biological transformations take place simultaneously rather than in a sequential manner. This shows that natural phenomena can be seen as massively parallel processes, since they occur above at least two independent levels of parallelism. The work by Cecilia and colleagues provides an example of a formal language that highlights the massive parallelism typical of biology. Based on this language and on CUDA, the computing engine running in NVIDIA graphics processing units, they present their software setting. Revolutionary biology demanded revolutionary computing. Great goals have been achieved since the SPARCcenter 2000 was used to assemble the genome of Haemophilus influenzae in 1995. That was the last model of a computer generation limited by an architecture capable of addressing only 2 GB of RAM. Modern processors are almost 200 times more powerful. The cost of storage has also dropped dramatically over the years. In 1992, 1 TB of disk space cost 1 million dollars; in 2010, the cost has been reduced to nearly 100 dollars. Furthermore, while in 1992 10 MB/s was the typical speed in networks, today gigabit network interfaces are very common and specialized processors are affordable for all. Prandi and Dematté provide an overview of how Systems Biology has been affected by such circuits as a whole. In this direction, works of Richmond, Walker, Coakley and Romano and that of Bako are particularly noteworthy. The former makes use of the flexible large-scale agent modeling environment called ‘FLAME’ to simulate cellular-level processes on graphics processing units, while the latter demonstrates that artificial spiking neural networks built to resemble the biological model encoding information in the timing of single spikes are capable of computing and learning clusters from realistic data making use of embedded soft-core microcontrollers. Thus, the very nature of large-scale computing has changed from systems relying on one or a few powerful custom-designed processors to scalable parallel systems or farms of computer. Any automatic procedure is approaching an alternative parallel implementation; and several, not trivial, ad hoc synchronization policies are sometimes coming up to support their final deployment. The very revolution lies in the fact that complex problems can now be broken down into sets of smaller jobs and can be executed on high-performance machines nodes, even remotely available, through easy-to-use access interfaces. The paper entitled Estimating the divisibility of complex biological networks by sparseness indices is an exemplary attempt to find a rationale behind the exhausting quest for ‘smaller and clever’ computational subunits. Hence, to cut a long story short, I am of the view that there are many-sided horizons where researchers are going. If I were to mirror their needs and dreams, I believe they can be reached only by crossing the rivers of accessibility, feasibility; quickness and reliability, important properties of any future framework. This issue tries to provide the reader with some clue in finding their own way. I hope you will enjoy reading about this and that this issue has done the job of exciting your interest and has been particularly helpful in encouraging you to find a common and long-run target on this path.

In order to understand the complex relationships among the components of biological systems, network models have been used for a long time. Although they have been extensively used for visualization, data storage, structural analysis and simulation, some computational processes are still very inefficient when applied on complex networks. In particular, any parallel simulation technique requires a network previously divided into a number of clusters in numbers equal to that of the available processors. At the same time, let maximally disconnected clusters be chosen in order to minimize extra-communication overhead and to optimize the overall computational efficiency. Obtaining such a disconnection becomes a computationally hard problem when disconnection conditions are complex in themselves, like in the case of parallel simulation. Before applying any clustering method, topological indices might contribute to give an a priori insight about the divisibility of a network. Here we present a class of them, the sparseness indices. As particular topological indices provide either local or global quantification of network structure, they can help in identifying locally dense, but globally sparsely connected subgraphs.

Biological systems are characterised by a large number of interacting entities whose dynamics is described by a number of reaction equations. Mathematical methods for modelling biological systems are mostly based on a centralised solution approach: the modelled system is described as a whole and the solution technique, normally the integration of a system of ordinary differential equations (ODEs) or the simulation of a stochastic model, is commonly computed in a centralised fashion. In recent times, research efforts moved towards the definition of parallel/distributed algorithms as a means to tackle the complexity of biological models analysis. In this article, we present a survey on the progresses of such parallelisation efforts describing the most promising results so far obtained.

MOTIVATION: Compartments and membranes are the basis of cell topology and more than 30% of the human genome codes for membrane proteins. While it is possible to represent compartments and membrane proteins in a nominal way with many mathematical formalisms used in systems biology, few, if any, explicitly model the topology of the membranes themselves. Discrete stochastic simulation potentially offers the most accurate representation of cell dynamics. Since the details of every molecular interaction in a pathway are often not known, the relationship between chemical species in not necessarily best described at the lowest level, i.e. by mass action. Simulation is a form of computer-aided analysis, relying on human interpretation to derive meaning. To improve efficiency and gain meaning in an automatic way, it is necessary to have a formalism based on a model which has decidable properties. RESULTS: We present Cyto-Sim, a stochastic simulator of membrane-enclosed hierarchies of biochemical processes, where the membranes comprise an inner, outer and integral layer. The underlying model is based on formal language theory and has been shown to have decidable properties (Cavaliere and Sedwards, 2006), allowing formal analysis in addition to simulation. The simulator provides variable levels of abstraction via arbitrary chemical kinetics which link to ordinary differential equations. In addition to its compact native syntax, Cyto-Sim currently supports models described as Petri nets, can import all versions of SBML and can export SBML and MATLAB m-files. AVAILABILITY: Cyto-Sim is available free, either as an applet or a stand-alone Java program via the web page (http://www.cosbi.eu/Rpty_Soft_CytoSim.php). Other versions can be made available upon request.

This paper presents a methodology for analyzing and classifying proteins identified in biological samples. In particular, such methodology consists in normalizing and classifying quantity and quality of proteins identified by using tandem mass spectrometry. A case study is considered and a classification experiment for protein discriminant is also reported

The combined use of mass spectrometry and data mining is a novel approach in proteomic pattern analysis for discovering novel biomarkers or identifying patterns and associations in proteomic profiles. Data produced by mass spectrometers are affected by errors and noise due to sample preparation and instrument approximation, so different preprocessing techniques need to be applied before analysis is conducted. We survey different techniques for spectra preprocessing, and we present a first design of a software tool that allows the preprocessing, management and analysis of mass spectrometry data on the Grid.