Programme

Preliminary Program

A booklet containing an outline of the WiN COSTNET event, program, abstract, participants lists and useful information, can be found on Luisa Cutillo's github page here.

Day 1 - Wed 27 February - MALL 2, Level 8, School of Mathematics
9:00am	Arrival and Registration - Registration Desk Level 9, School of Mathematics
9:45am	Welcome by Jeanine Houwing-Duistermaat and Luisa Cutillo - University of Leeds
10:00am	Keynote: Veronica Vinciotti - Brunel University, London
11:00am	Coffee Break
11:30am	Pariya Behrouzi, Wageningen University, NL Title:- Construction of High-resolution Linkage Maps Using Discrete Graphical Models
12:00 noon	Rafiazka M. Hilman, Central European University (CEU) Title:- The Dynamics of Mass and Elite in Dutch Dividend Tax Discourse
12:30pm	Lunch at the refectory
2:00pm	Keynote: Gesine Reinert - University of Oxford
3:00pm	Breakout 1 - Challenges in Biological Networks
3:30pm	Coffee Break
4.30pm	Report back and plenary
7pm	Social Dinner

Day 2 - Thur 28 February - Morning session at Worsley Seminar Room 8.34a
9:30am	Arrival and Registration, Level 9, School of Mathematics
9:45am	Sustainable garden tour and walk to Worsley Seminar Room 8.34a (via Garstang building)
10:00am	Keynote: Claudia Angelini - IAC-CNR Italy
11:00am	Coffee Break
11:30am	Marija Mihova, Ss. Cyril and Methodius University, FINKI, Skopje Title:- Efficient Algorithm for Finding all Minimal Path Vectors in Two-terminal Flow Network
12:00 noon	Martin Lopez Garcia, Department of Applied Mathematics, School of Mathematics, University of Leeds, UK. Title:- On the exact analysis of stochastic epidemic processes on networks
12:30pm	Lunch at the refectory - Afternoon session at LT E, Chemistry West Block
2:00pm	Keynote: Luisa Cutillo - University of Leeds
3:00pm	Valeria Policastro, IAC Istituto per le Applicazioni del Calcolo "Mauro Picone", CNR Naples, Italy Title:- ROBIN: an R package for validation of community robustness
3:30pm	Coffee Break
4pm	Breakout 2 - Identifying potential solutions
5pm	Report back and plenary

Day 3 - Fri 1 March - MALL 2, Level 8, School of Mathematics
9:30am	Keynote: Jeanine Houwing-Duistermaat - Department of Statistics, School of Mathematics, University of Leeds
10:30am	Rebeka O. Szabo, CEU Title:- The micro-dynamic nature of team interactions
11:00am	Coffee break
11:30am	Keynote: Marta Milo - University of Sheffield
12:30pm	Denise De Gaetano, MCAST Title:- Extracting the Value of Big Data
1:00pm	Arief Gusnato, School of Mathematics, University of Leeds Title:-Transcriptomic and Genomic Networks
1:30pm	Lunch&Posters - Reading Room 9.31, Level 9, School of Mathematics
2:30pm	Poster Session Discussion
3.30pm	Coffee Break and Close

After the notification of acceptance, please purchase your lunches and social dinner options on our online store here.

ABSTRACTS

Keynote Veronica Vinciotti - Network inference in genomics under censoring

Regularized inference of networks using graphical modelling approaches has seen many applications in biology, most notably in the recovery of regulatory networks from high-dimensional gene expression data. Under an assumption of Gaussianity, the popular graphical lasso approach provides an efficient inferential procedure under L1 sparsity constraints. In this talk, I will focus on a latest extension to censored graphical models in order to deal with censored data such as qPCR expression data. We propose a computationally efficient EM-like algorithm for the estimation of the conditional independence graph and thus the recovery of the underlying regulatory network. Similar techniques can be used also in the context of multivariate regression where censored outcomes are to be predicted from a set of predictors. Efficient inferential procedures are presented in the high-dimensional case and pave the way for the development of more complex models that integrate data from different sources and under different mechanisms of missingness.

Keynote Claudia Angelini - An overview on penalized network regression approaches.

In this talk we will briefly review the main concepts and problems that arise when analyzing high dimensional data, then we describe recent approaches based on network penalized regression. As an illustrative example, we describe a novel method that combines variable screening and penalized network-based Cox-regression models for the identification of high- and low-risk groups in breast cancer and the selection of potential biomarkers. More in general, we illustrate most recent results and open challenges of network penalized approaches in the context of omic data analysis and integration.

Keynote Gesine Reinert - Anomaly detection in networks.

Detecting financial fraud is a global challenge. This talk will mainly focus on financial transaction networks. In such networks, examples of anomalies are long paths of large transaction amounts, rings of large payments, and cliques of accounts. There are many methods available to detect specific anomalies. Our aim is to detect unknown anomalies. To that purpose we use a strategy with derives features from network comparison methods and spectral analysis, and then apply a random forest method to classify nodes as normal or anomalous. We test the method on synthetic data which we generated, and then on synthetic data without us having had access to the ground truth.
This talk is based on joint work with Andrew Elliott, Mihai Cucuringu, Milton Martinez Luaces, Paul Reidy.

Keynote Luisa Cutillo - Main challenges in Networks community structure validation.

High throughput technologies have led to an increased availability of data and to the need for novel statistical tools. Biological networks provide a mathematical representation of patterns of interaction between appropriate biological elements. We propose a novel approach to compare community structures in different networks. During this seminar we will try to address some open questions: How can we compare two (or more) networks and their community structures? Can we use Network Enrichment Analysis tools to do this? Is it an advantage to integrate metadata to infer communities?

Keynote Jeanine Houwing-Duistermaat - Data integration using Network and Partial Least Square methods .

The availability of large omics datasets in epidemiological and clinical studies provides many opportunities for research in statistical bioinformatics. The hope is that the abundance of information will provide better understanding of underlying disease mechanisms and accurate prediction models enabling patient targeted screening and treatment. Statistical challenges are to deal with data wrangling, heterogeneity across omic datasets, high dimensionality, data integration and the presence of high correlation within and between datasets (Morris et al, 2017; Houwing-Duistermaat et al, 2017). In this talk I will present Partial Least Squares (PLS) and Network methods for data integration and dimension reduction when analysing several omics datasets simultaneously. The methods will be illustrated by analysis of glycomic datasets and of metabolomics and gene expression in relation with Body Mass Index.

Keynote Marta Milo - Bring Mathematics into Biology: Past, Present and Future Impact on Health

Last decade has seen a massive increase of data production in science. Particularly in the biomedical field, data has grown exponentially thanks to the development of technologies like next generation sequencing and high-throughput quantitative assays. The information that this data contains is only partially uncovered to this date, but the impact that it has on human progression and well-being is already very clear.
Despite the ability to process large amount of data and to quantify fine details of biological processes, the costs, the time to perform such experiments and mainly the complexity of the systems remain in some cases still very prohibitive. For this reasons the use of mathematics to study complex systems in its entirety, looking at how they interacts, is having a great impact in current biology and healthcare.
A variety of statistical, probabilistic and optimisation techniques methods, like machine learning techniques, that allows to “learn” from the available data, to detect hidden patterns from large, noisy and complex datasets, is particularly suitable for application in medicine. In this talk I will present examples of using machine learning techniques for a variety datasets from medical and biological problems and what are the advantages and disadvantages of this approach. I will also give examples when these techniques enabled to discover informative knowledge from a large complex system in the presence of small number of samples. Finally I will discuss how we use Machine Learning today for analysis of single-cell sequencing data and how we can use it for future more complex datasets generated integrating data from different sources.