Here you can find the abstracts and, as soon as available, the slides and the videos for all the talks of the , sorted in the order they have been presented. Remember to refresh the page to see the latest additions.

Rubem P. Mondaini (Federal University of Rio de Janeiro, Brazil), Invited speaker: The BIOMAT Consortium
 Abstract: The BIOMAT Consortium is an international non-profit association of scientific faculty members of universities and research institutions worldwide, their scientific research students as well as other interested scientific practitioners on the areas of Mathematical Biology, Biological Physics and the generic mathematical modelling of biosystems. The fundamental mission of the BIOMAT Consortium is to promote the scientific exchange of knowledge in the interdisciplinary areas of Mathematical and Biological Sciences as well as the enhancement of multidisciplinary scientific work on developing countries worldwide. Its President, Rubem P. Mondaini, will briefly introduce the Consortium and its activities.
Slides:
Video:

Mario Nicodemi (University of Naples “Federico II”, Italy), Invited Speaker: Chromatin 3D architecture: mechanisms and function
 Abstract: I discuss our recent work to understand the mechanisms whereby human chromosomes are folded and regulated in the nucleus of cells. Our genome has a complex 3D organisation serving vital functional purposes as, for instance, genes need to contact distal DNA regulatory sequences to be activated. Yet, it is unknown how the system self-organises. First, I overview our recent experimental advances to measure the 3D structure of chromosomes. Next, I discuss our theories showing that chromosomal architecture is controlled by mechanisms of polymer phase transition. Those theories have been confirmed by recent experiments, in particular, on their predictions on how genetic mutations result in diseases such as congenital disorders and cancer.
Slides:

Chiara Lanzuolo (ITB-CNR and INGM, Italy), Invited Speaker: Novel technological approaches to explore nuclear architecture dynamics in biological processes and disease
 Abstract: Chromatin within the cell nucleus has a complex structure that is fundamental for genome function. The characteristic of the cells, their plasticity and the ability to respond to the environmental stimuli depend on the chromatin shape and remodelling properties. In recent years, what is emerging is that, besides the plasticity of the chromatin fundamental for fine-regulated process, the nuclear architecture can also influence important cellular processes and is a hallmark of the healthy cell. Alterations in chromatin and/or nuclear structures are associated with developmental defects, genetic diseases and cancer, while its proper conformation is a hallmark of healthy cells. My group is devoted in understanding how the genome folding occur in the nuclear space finding the right orientation and nuclear position and how this conformation is then maintained or regulated in dynamic physiological processes in health and in disease. We started studying epigenetic factors known to play a key role in genome folding and function, the Polycomb group (PcG) of proteins. We described for the first time a functional and evolutionary conserved crosstalk between the nuclear Lamin A/C and the PcG proteins; this being required for the maintenance of the PcG repressive functions. In two recent works, we show how mutation of Lamin A can generate a dysfunctional Polycomb program leading to a defect in cell identity maintenance in laminopathies, human disorders caused by mutations in Lamin A/C gene. We are now committed in understanding the role of PcG/LaminA interplay in aging and cancer. In parallel we developed a new high-throughput sequencing technique, the Sequential Analysis of MacroMolecule accessibilitY (SAMMY-seq) to study different levels of chromatin solubility. This technology, applied in several primary tissues and cell types, will be instrumental to investigate the role of chromatin conformation in pathology insurgence and/or progression.
Slides:
Video:

Laura Antonelli (ICAR-CNR, Italy): Identification and analysis of intranuclear protein patterns in fluorescence microscopy images
 Abstract: Advances in development of genetically encoded fluorescent proteins and in digital imaging has led to the rapid evolution of live-cell imaging methods. These methods are being applied to address biological questions, in particular the identification of the intranuclear protein pattern can help the analysis of specific processes, such as DNA repair, DNA integration, and chromatin folding. Here, we present an efficient tool that implements mathematical algorithms to detect the pattern of the Polycomb Group (PcG) of proteins in high resolution fluorescent image cell stacks. Our tool is composed of an automatic segmentation algorithm combining the globally convex Chan-Vese model and a classification method, to segment nuclei regions and detect intranuclear PcG areas. Then a 3d reconstruction step of nuclei and proteins is performed, followed by a set of algorithms designed to explore the 3d structure in order to produce a quantitative analysis of nuclei and proteins, and to evaluate the intranuclear positioning of the PcGs. The 3D reconstruction of several cell image datasets, coming from different cell types and corresponding to different fluorescent labels, has showed that intranuclear positioning of PcG bodies is evolutionarily conserved, being horizontally coplanar and excluded from the nuclear periphery.
Slides:

Panos M. Pardalos (University of Florida, USA, and LATNA, National Research University, Russia), Invited Speaker: Artificial Intelligence in Biomedicine
 Abstract: Artificial Intelligence (AI) has been a fundamental component of many activities in biomedicine in recent years. In the first lecture we summarize some of the major impacts of AI tools in biomedicine and discuss future developments and limitations. In the second lecture we present details on our recent research on a kidney project.
Slides:

Maurizio Giordano (ICAR-CNR, Italy): Graph embedding for biological networks
 Abstract: In Computational Biology, several research efforts are devoted to the development of efficient techniques to process and analyse large amount of biological data and their relations. In this context, one of the main problems is how to reduce the complexity of biological networks, represented as graphs, through projections and/or transformation into a more manageable data space. Graph Embedding (GE) techniques pursue this scope, by translating large and complex graphs into a reduced vector space called ''latent space''. Several GE techniques have been proposed in literature, which rely on different approaches, such as graph kernels, graph traversal and random walks, and deep learning. In this talk we present a new GE method, called Netpro2vec. It exploits graph node proximity information to transform graphs into textual documents while preserving their significant structural properties. Netpro2vec relies on an NLP learning model to extract, from each document-based graph, the meaningful features in terms of vectors, i.e. the ''embeddings''. Such a new graph representation can be used for different machine learning tasks, such as, unsupervised clustering and supervised classification of graphs. The advantage of Netpro2vec is that it provides efficient embeddings completely independent from the task and nature of the data.
Slides:

Mario Manzo (University of Naples "L'Orientale", Italy): Whole graph embedding: robustness and vulnerability
 Abstract: Graph embedding techniques are becoming increasingly common in many fields ranging from scientific computing to biomedical applications and finance. These techniques aim to automatically learn low-dimensional node representations for a variety of network analysis tasks. In literature, several methods (e.g., random walk-based, factorization-based, and neural network-based) show very promising results in terms of their usability and potential. Despite their spreading diffusion, little is known about their reliability and robustness, particularly when applied to the real world of data, where adversaries or malfunctioning/noising data sources may supply deceptive data. The vulnerability emerges mainly by inserting limited perturbations on the input data when these lead to a dramatic deterioration in performance. To this end, an analysis of different adversarial attacks in the context of whole-graph embedding is proposed. The attack strategies involve a limited number of nodes based on the role they play in the graph. The study aims to measure the robustness of different whole-graph embedding approaches to those types of attacks, when the network analysis task consists in the supervised classification of whole-graphs. Extensive experiments carried out on synthetic and real data provide empirical insights on the vulnerability of whole-graph embedding models to node-level attacks in supervised classification tasks.
Slides:
Video:

Marta Milo (AstraZeneca, UK), Invited speaker: Applications of Data Science to define interactions from gene network to drug combinations
 Abstract: The advances in understanding disease complexity by modelling data integration offer the opportunity to explore “omics” datasets with data science approaches. In this talk I will give an example of how we can use transcriptomic analysis to reveal a robust gene expression signature in low-level gene expression conditions and an example on how to use generalised linear models to extract low frequency signalling pathways in an oncology setting. Finally, the talk will touch on the ability of identifying drug combinations. The use the molecular characterisation of diseases and the development of technologies for high-throughput pharmacology screenings, generated a pressing need to identify efficacious drug combinations, able to achieve more effect with less toxicity and thus improving prognosis and life-quality. I will explore the space of drug combinations and the challenges that it offers highlighting how we can use Data Science and Scientific Computing to innovate the methodologies to predict and optimise drug combinations.
Slides:
Video:

Diego Di Bernardo (Telethon Institute of Genetics and Medicine, Italy), Invited speaker: An integrated microscopy and microfluidics platform for real-time automatic control of biological process in living cells
 Abstract: By combining tools from Control Engineering and Synthetic Biology (cybergenetics), we propose a simple and cost-effective microfluidics-based platform to precisely regulate gene expression and signaling pathway activity in cells by means of real-time feedback control. The platform performs the following steps: (1) the output of interest is measured in cells. This is usually a fluorescent protein whose level is observed with a time-lapse fluorescence microscope; (2) protein level is quantified in individual cells from microscopy images with a custom image processing algorithm; (3) a computer implementing the control strategy computes the control input needed to minimise the difference between the target level and the actual population-averaged fluorescence intensity across the cells. We will show applications of the platform to automatically control protein expression and cell cycle in yeast cells.
Slides:
Video:

Diego Romano (ICAR-CNR, Italy): Tracking yeast cells in phase-contrast microscopy image sequences by exploiting GPUs
 Abstract: When information and measures obtained from sequences of microscopic images are subject to time constraints, suitable fast algorithms must be implemented to process the whole data set. In this case, we deal with sequences of images obtained from time-lapse microscopy to detect single yeast cells in a microfluidics chip over time. The underlying idea consists of determining a minimum cost configuration for each couple of frames, which can be expressed by setting up and solving a linear programming (LP) problem. Laboratories seldom have the opportunity to use HPC hardware for such intent. For this reason, we propose an efficient GPU-parallel software implemented in CUDA and based on the simplex method, a standard tool for solving LP problems. Thanks to minimal thread divergence and data locality, we recorded a promising speedup over the CPU version on authentic images sequences, making our GPU-parallel software suitable for real-time applications.
Slides:
Video:

Annabella Astorino (ICAR-CNR, Italy): Linear and nonlinear approaches for Multiple Instance Learning
 Abstract: Multiple Instance Learning (MIL) is a variant of traditional supervised learning that has received a considerable amount of attention due to its applicability to real-world problems such as drug activity prediction and image classification. The main difference with the traditional supervised learning scenario is in the nature of the learning examples. In fact, each example is not represented by a fixed-length vector of features but by a bag of feature vectors called instances. The classification labels are only provided for entire training bags whereas the labels of the instances inside them are unknown. The task is to learn a model that predicts the labels of the new incoming bags together the labels of the instances inside them. In this work we focus on the binary case, characterized by two types of bags and two types of instances, where a common assumption consists in considering a bag positive if it contains at least a positive instance and negative if it contains only negative instances. Starting from this assumption and initially inspired by a well-established SVM type approach, we present some spherical and polyhedral approaches. The idea is to generate a non linear separation surface of spherical and polyhedral type such that, for each positive bag, at least one of its instances is inside the sphere or the polyhedron and all the instances of each negative bag are outside. Numerical results are presented on some test problem drawn from the literature.
Slides:
Video:

Giovanni Schmid (ICAR-CNR, Italy): Genomic databases: prospects and challenges
 Abstract: Huge DBMSs storing genomic information are being created and engineerized for doing large-scale, comprehensive and in-depth analysis of human beings and their diseases. This paves the way for significant new approaches in medicine, but also poses major challenges for storing, processing and transmitting such big amounts of data in compliance with recent regulations concerning user privacy. This talk discusses computational and security issues for such DBMS, introducing a new efficient and secure architecture.
Slides:
Video:

Ferdinando Montecuollo (University of Campania “Luigi Vanvitelli”, Italy), Invited speaker: Building genomic databases with ER-index
 Abstract: The ER-index is an open-source C++ tool designed to handle an encrypted genomic database. Thanks to a multi-user and multiple-keys encryption model, a single ER-index can store the sequences related to a large population of individuals so that users may perform search operations directly on compressed data and only on the sequences to which they were granted access. This presentation illustrates the features and the practical usage of this new index through a demo showing how to build a genomic database.
Slides:
Video:

Mario R. Guarracino (University of Cassino and Southern Lazio, Italy): Farewell
Audio: