Annalisa Barla

• Parameter Space Exploration within Dynamic Simulations of Signaling Networks, Cristina De Ambrosi, Annalisa Barla, Lorenzo Tortolina, Nicoletta Castagnino, Raffaele Pesenti, Alessandro Verri, Alberto Ballestrero, Franco Patrone, Silvio Parodi, Mathematical Biosciences and Engineering, 2013
• Knowledge Driven Variable Selection (KDVS) - a new approach to enrichment analysis of gene signatures obtained from high-throughput data, Grzegorz Zycinski, Annalisa Barla, Margherita Squillario, Tiziana Sanavia, Barbara Di Camillo, Alessandro Verri, Source Code for Biology and Medicine, 2013
• Dictionary learning improves subtyping of breast cancer aCGH data, Salvatore Masecchia, Annalisa Barla, Saverio Salzo, Alessandro Verri, Proc IEEE EMBS, 2013
• A dictionary learning based method for aCGH segmentation, Salvatore Masecchia, Saverio Salzo, Annalisa Barla, Alessandro Verri, Proceedings of ESANN2013, 2013
• Multi-Output Learning via Spectral Filtering, Luca Baldassarre, Lorenzo Rosasco, Annalisa Barla, Alessandro Verri, Machine Learning, 2012

• Molecular characterization of neuro-degenerative diseases
• Prior Knowledge Integration in high-dimensional molecular data
• Knowledge discovery from biological data
• Determining gene signatures from high-throughput data

• Modelli e metodi computazionali nello studio della fisiologia e patologia di reti molecolari di controllo in ambito oncologico.

My research mainly focuses on statistical learning techniques set in the context of regularization methods and applied to computational biology.

The analysis of biological data is characterized by a large amount of features representing each given example (high-dimensionality) and a relatively small number of samples. Traditional statistical tools were developed and studied to adapt in the opposite scenario, where the samples outnumber many times the variables, so different approaches have to be explored.

My main interests, from the algorithmic viewpoint are:

• Supervised learning/classification methods
• Kernel Engineering
• Feature Selection
• Vector valued regression/classification
• Clustering techniques

From the methodological viewpoint, I am interested in studying statistical analysis algorithms which guarantee robustness and unbiased results, such as regularization techniques combined with validation methods to properly select the parameters:

• significance assessment - complete validation
• model selection

My other webpage is here and also here.