André Veríssimo

Studied how to improve sparse generalized linear models by introducing network-based centrality regularization in cancer datasets. The methods that we developed allow for the discovery of novel genes of interest as well as an improvement on the models itself and how the features are selected.

Developing of sparse models for high-dimensional multi-omics patient data and model-based outlier detection. This was implemented as part of work packages WP4.1 and WP4.2

Developed software to automate the execution of multiple BLAST queries against multiple databases as part of work package WP8 of BACHberry EU Project FP7-613793

Coordinator at GASTagus responsible for the internal collaborative platforms, communication and financial management. GASTagus is a non-profit organization that manages over 130 volunteers during all-year round, preparing them for a one month volunteer project in a African country or Brasil.

Researcher at INESC-ID in Knowledge Discovery and Bioinformatics group. Developed a web application, where a rich set of dynamic models can be fitted to experimental time-series data with data hierarchical data management. Several dynamic and algebraic models are already implemented and run in Matlab in the back-end.

Consultant at Deloitte’s Consulting division, in Manufacturing, Consumer Business, Energy and Resources industries on the Enterprise Applications service line. Professional experience in the logistics and operations processes. Those projects, took place mostly in manufacturing and consumer business organizations, across multiple business sectors and countries, such as, Portugal and Angola.

Junior consultant at Karma consulting in the public sector service line. Professional experience in logistics and information systems design working with Portugal's public sector entities as a software engineer specialized in Java development

Studied how to improve sparse generalized linear models by introducing network-based centrality regularization in cancer datasets. The methods that we developed allow for the discovery of novel genes of interest as well as an improvement on the models itself and how the features are selected.