Conference Program

The program of the CIBB 2018 International Conference is displayed here.

The Conference Dinner will take place at:

Casa da Cerca – Centro de Arte Contemporânea

Rua da Cerca

2800 – 050 Almada

Coordenadas GPS
38° 41' 0.820" N    9° 9' 33.502" W



Professor Alberto Paccanaro,  University of London, U.K.

Professor Alexandra Carvalho, Universidade de Lisboa, Portugal

Professor Benoit Liquet, University of Pau, France

Professor Fernando L. Ferreira, Universidade Nova de Lisboa, Portugal

Professor Veronica Vinciotti, Brunel University London, UK


Answering questions in biology and medicine by making inferences on networks


An important idea that has emerged recently is that a cell can be viewed as a complex network of interrelating proteins, nucleic acids and other bio-molecules.  At the same time, data generated by large-scale experiments often have a natural representation as networks such as protein-protein interaction networks, genetic interaction networks, co-expression networks. From a computational point of view, a central objective for systems biology and medicine is therefore the development of methods for making inferences and discovering structure in biological networks possibly using data which are also in the form of networks. In this talk, I'll present novel computational methods for solving biological problems which can all be phrased in terms of inference and structure discovery in large scale networks. These methods are based and extend recent developments in the areas of machine learning (particularly semi-supervised learning and matrix factorization), graph theory and network science. I will show how these computational techniques can provide effective solutions for: 1) quantifying similarity between heritable diseases at molecular level using exclusively disease phenotype information; 2) disease gene prediction; 3) drug side-effect prediction.


Model selection for temporal biomedical data


Human health care is changing rapidly, pressing the development of machine learning techniques for automatic diagnoses and prognosis, as well as personalized therapies for individual patients. The emerging availability of temporal data, namely via electronic medical records, is triggering this line of research. One of the main problems is to model the dynamic process underlying the data evolution. We detail how to learn efficiently Markovian data, when the dependencies can be expressed as a dynamic Bayesian network. We follow a score-based approach, and guarantee that the learned model is optimal according to several model selection criteria. Finally, we address the problem of early classification, which is essential in time-sensitive applications, such as personalized therapies.


A Unified  Regularized Group PLS Algorithm Scalable to Big Data.  Application on genomics data


Partial Least Squares (PLS) methods have been heavily exploited to analyse the association between two blocs of data. These powerful approaches can be applied to data sets where the number of variables is greater than the number of observations and in presence of high collinearity between variables. Different sparse versions of PLS have been developed to integrate multiple data sets while simultaneously selecting the contributing variables. Sparse modelling is a key factor in obtaining better estimators and identifying associations between multiple data sets. The cornerstone of the sparsity version of PLS methods is the link between the SVD of a matrix (constructed from deflated versions of the original matrices of data) and least squares minimisation in linear regression. We present here an accurate description of the most popular PLS methods, alongside their mathematical proofs. A unified algorithm is proposed to perform all four types of PLS including their regularised versions. Our methods enable us to identify important relationships between genomic expression and cytokine data from an HIV vaccination trial. We also proposed a new methodology by accounting for both grouping of genetic markers (e.g. genesets) and temporal effects.  Finally, various approaches to decrease the computation time are offered, and we show how the whole procedure can be scalable to big data sets.


Ethics and our moral in research, let’s think about it!


As researchers, it is our will is to pursue knowledge, to contribute to society and to open new roads for the Future. Ethics is a theme always present in our minds but probably remains outside the central concerns of researchers while main subjects are developed. Sometimes we come across a formal consent or an ethics statement seen mostly as a bureaucratic task. However, lately with the so called exponential technologies, we find ourselves duelling with a variety of controversial questions resulting from the different branches of artificial intelligence as those applied to self-driving cars’ decisions the exposure of privacy and Decision support systems in medicine, etc. Some are arguing that risks become clear and, one of this days, we may face a singularity and, eventually, becoming too late to stop. This short talk aims to rise some questions about present ethical issues aiming at promoting the intervention and discussion among participants at this Conference.


Sparse graphical models in genomics: an application to censored qPCR data


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. Various extensions to the standard graphical lasso approach have been proposed, such as dynamic and hierarchical graphical models. In this talk, I will focus on a latest extension to censored graphical models in order to deal with censored data such as qPCR 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.