Research Activity

Bioinformatics research activities are characterized by the development and application of machine learning methods and algorithms to the analysis of high-throughput bio-molecular data. Schematically, the research activity may be set out in five main areas, in turn articulated in several sub-areas:

Bioinformatics

A) Ensemble based integration of heterogeneous
biomolecular data.

A1. Analysis, development and application of data fusion systems for gene function prediction.
A2. Analysis, development and application of data fusion systems for the prediction of the expression patterns of co-regulated genes.

B) Structured output learning.

B1. Ontology-based hierarchical classification of genes and proteins.

C) Unsuprvised learning and stability based clustering.

C1. Ensemble clustering methods for the analysis of patterns in bio-molecular data

D) Analysis, development and application of supervised
machine learning methods in bioinformatics.

D1. Development of machine learning methods for the identification of unannotated genes and novel splicing isoforms in eukaryotes.
D2. Development of machine learning methods for the assessment of the protein-coding potential of mRNAs in high-throughput sequencing projects.
D3. Development of network-based machine learning methods for drugs repositioning.

E) Other bioinformatics research activities.

E1. DNA microarray data analysis for the discovery of gene networks related to Human Acute Myeloid Leukaemia stem cells.
E2. Ontology driven gene selection methods for the discovery of functional classes of genes related to a specific phenotype