The Systems Engineering in the Life Sciences area is actively pursuing research in the fields of computational and systems biology, biostatistics and machine learning with applications to biological and clinical problems. We are currently collaborating with other groups and institutions (see People and Projects) and always try to provide accompanying Software to our methods.

We are currently addressing the following main topics:

1) Dynamic modeling, identification and optimization of biological systems

modelglucose We are interested in modeling complex physiological systems, using several methodologies such as ordinary differential equations and time-series analysis. The applications include metabolic networks modeling and optimization, HIV infection and cell growth.

The study of metabolic networks, the set of chemical processes and reactions occurring in cells, has undergone a considerable increase in recent years. The growing importance of this research is a reflection of its expected impact on several areas, such as food and pharmaceutical industry, biotechnology and medicine. In particular, modeling the dynamic behavior of these networks and their optimization for the production of given compounds might be regarded in the context of theoretical and computational (dry lab) systems biology, an emerging field that uses a global and integrative perspective to capture the behavior of complex living organisms. We currently address methods for modeling the dynamics of metabolic networks using systems of non-linear differential equations. Several difficulties are associated with model identification and parameter estimation from experimental time-series, which usually leads to hard optimization problems. A case study on the glycolytic pathway in bacteria illustrates the success of these methodologies for the in silico simulation and prediction of glucose metabolism. Another key application of optimization in metabolic engineering is the estimation of the required changes in the network in order to maximize given reaction fluxes, which can be formulated as a mixed-integer linear programming problem. Recently, the relations between multi-objective optimization and metabolic engineering of cell communities are being further explored, strengthening the need of optimization strategies in systems biology.

Projects: BacHBerry, HIVCONTROL, DynaMo
Collaborators: João M. Lemos, Ana R. Neves, Marie-France Sagot
Selected references:

  • A Hartmann, JM Lemos, RS Costa, J Xavier and S Vinga. Identification of Switched ARX Models via Convex Optimization and Expectation Maximization. Journal of Process Control. (Accepted).
  • Vinga S, Neves AR, Santos H, Brandt BW, and Kooijman SALM (2010) Subcellular metabolic organization in the context of Dynamic Energy Budget and Biochemical Systems theories. Philosophical Transactions of the Royal Society B. 2010 Nov 12;365(1557):3429-42.
  • Vilela, M., Chou, I. C., Vinga, S., Vasconcelos, A. T. R., Voit, E. O., and Almeida, J. S. (2008). Parameter optimization in S-system models. BMC Systems Biology, 2, 35.

2) Statistical and machine learning methods for longitudinal clinical data analysis

survival-stat We are interested in Data Science applications to clinical problems by developing statistical and machine learning methods and tools. Some problems being addressed include: clustering short time-series data, classification and feature extraction, survival analysis of bone metastatic patients and alternative splicing events, genotype/phenotype associations, and, more generally, adaptive clinical decision support systems for personalized medicine.

Projects: PERSEIDS, SOUND, CancerSys, InteleGen
Collaborators: Alexandra M. Carvalho, Nuno Barbosa-Morais, Arlindo Oliveira, Sara Silva
Selected references:

  • Caldas J, Vinga S (2014) Global Meta-Analysis of Transcriptomics Studies. PLoS ONE 9(2): e89318. doi:10.1371/journal.pone.0089318
  • Tenazinha N, and Vinga S (2011) A Survey on Methods for Modeling and Analyzing Integrated Biological Networks IEEE/ACM Transactions on Computational Biology and Bioinformatics (TCBB) 8(4), pp. 943-958, Jul./Ago. 2011.

3) Alignment-free sequence analysis and information theory

cgr Alignment-free methods for biological sequence analysis and comparison have emerged as a natural framework to address the challenges of understanding the patterns and properties of biological sequences. We have been developing methods to analyse DNA and proteins using Information theory concepts.

Collaborators: Jonas S Almeida
Selected references:

  • Vinga, S. (2014). Information theory applications for biological sequence analysis. Briefings in Bioinformatics, 15(3), 376-389. doi: 10.1093/bib/bbt068
  • Vinga S, Carvalho AM, Francisco AP, Russo LMS, Almeida JS. (2012) Pattern matching through Chaos Game Representation: bridging numerical and discrete data structures for biological sequence analysis. Algorithm Mol Biol 2012, 7:10.
  • Vinga, S., and Almeida, J. (2003). Alignment-free sequence comparison – a review. Bioinformatics, 19(4), 513-523.

 On going projects

  • PERSEIDS – Personalizing cancer therapy through integrated modeling and decision. Jun 2016 – Jun 2019. Financed by: FCT (Contract PTDC/EMS-SIS/0642/2014)
  • SOUND – Statistical multi-Omics UNDerstanding of Patient Samples. Sep 2015 – Aug 2016. Financed by: Horizon2020 (Contract 633974)
  • BacHBerry – BACterial Hosts for production of Bioactive phenolics from bERRY fruits. Nov 2013 – Oct 2016. Financed by EUFP7 (Contract 613793)

Selected past projects

  • CancerSys –  Multiscale modeling for personalized therapy of bone metastasis. Apr. 2014 – Mar. 2015 Sept. 2015. Financed by FCT (EXPL/EMS-SIS/1954/2013)
  • Intelegen -Pharmacokinetic/Pharmacogenetic modulation of HIV infection therapy by bayesian and artificial intelligence methods. Jun. 2013 – May 2015 Oct 2015. Financed by FCT (PTDC/DTP-FTO/1747/2012)
  • HIVCONTROLJan. 2010 to Dec. 2012 – Control based on dynamic modeling of HIV-1 infection for therapy design. Prime contractor: INESC-ID. Financed by: FCT (PTDC/EEA-CRO/100128/2008)
  • DynaMoSept. 2007 to Aug. 2010 – Dynamic modeling, control and optimization of metabolic networks. Prime contractor: INESC-ID. Financed by: FCT (PTDC/EEA-ACR/69530/2006)
  • PNEUMOPATHMar. 2010 to Feb. 2012 – A comprehensive dissection of pneumococcal-host interactions. Prime contractor: University of Leicester (Peter Andrew, PI). Financed by: EUFP7 (Contract 222983)
  • PneumoSySJan. 2010 to Dec. 2012 – A systems biology approach to the role of pneumococcal carbon metabolism in colonization and invasive disease. Prime contractor: ITQB-UNL. Financed by: FCT (PTDC/SAU-MII/100964/2008)
  • EnviGPApr. 2010 to Apr. 2013 – Improving Genetic Programming for the Environment and Other Applications. Prime contractor: INESC-ID (S. Silva, PI). Financed by FCT (PTDC/EIA-CCO/103363/2008)