Innovation Capabilities & Products

Morbidostat is a computer-controlled continuous culture device that automatically adjusts drug concentration to maintain constant growth inhibition in microbial cultures. As bacteria acquire mutations that give them resistance against drugs, they are able to tolerate higher drug concentrations and grow faster, thus removing selective pressure, the driving force of evolution. To compensate for this, morbidostat increases drug concentration sufficiently to keep bacteria at their original growth rate, therefore maintaining selective pressure over time. This system allows for data acquisition to model microbial evolution under antimicrobial stress, optimize biocide dosage strategies and develop highly antimicrobial-resistant strains used to test the performance of new biocides, among other applications.

Method described in Csendes, T., L. Pal, J.O.H. Sendin, J.R. Banga (2008) The GLOBAL Optimization Method Revisited. Optimization Letters, 2(4):445-454.

Software available upon request (email us if interested).

Assessment and development of protocols to identify the most problematic areas and products in terms of microbial contamination and parasitic infestation along the value chain of seafood products.

AMIGO2 is a MATLAB-based multiplatform toolbox designed to solve mathematical optimization problems at the core of systems biology, within the context of parametric model identification, the underlying hypothesis for model development and the optimal control of biological systems to synthetically achieve the desired behavior.

More information here.

Method described in: Exler, O., L.T. Antelo, J.A. Egea, A.A. Alonso and J.R. Banga (2008) A Tabu search-based algorithm for mixed-integer nonlinear problems and its application to integrated process and control system design. Computers & Chemical Engineering, 32(8):1877-1891.

Software available upon request (email us if interested).

Development of Deep Learning algorithms that allow automating fisheries monitoring processes and reducing time and costs compared with processing by human observers. The applications developed range from innovative systems for real-time remote electronic monitoring, which identify and quantify total catches of fishing vessels (e.g., iObserver), to new image recognition techniques that allow individually identifying fish and estimating population parameters. 

PREMER (Parallel Reverse Engineering with Mutual information & Entropy Reduction) is an open-source, multi-platform software tool for inferring network structures from data using information-theoretic measures. While this general purpose tool has been developed with biological networks in mind, it can be applied to other areas.

More information here.

The method is described in Schlater, M., J. A. Egea, J. R. Banga (2009) Extended ant colony optimization for non-convex mixed integer nonlinear programming. Computers & Operations Research 36(7): 2217-2229.

Software available upon request (email us if interested).

iObserver is an innovative monitoring device based on automated video monitoring coupled with artificial intelligence developments for visual recognition and quantification of the catches on board fishing vessels.

iObserver implements a continuous image recording system adaptable to different fishing vessels and deep learning algorithms to automatically identify and quantify catches on board in real time.

iObserver focuses mainly on developing algorithms for robust automatic species recognition and size estimation of fish transported on a conveyor belt. Trials have been performed on board Spanish oceanographic vessels and commercial vessels. With over 300 days at sea, iObserver was used in more than 1000 hauls and took more than 200,000 pictures, and 17 species have already been included in the system's catalogue.  

For further information, please contact us by e-mail.

BioPreDyn-Bench is a suite of benchmark problems for dynamic modeling in systems biology. Currently, it contains six challenging parameter estimation problems, which aspire to serve as reference test cases in this area. This set includes medium- and large-scale kinetic models of E.coli, S. cerevisiae, D. melanogaster, Chinese Hamster Ovary (CHO) cells and a generic signaling network. The level of description includes metabolism, transcription, signal transduction and development.

More information here.