The research activities focus on elucidating the molecular mechanisms underlying human genetic diseases, mendelian and multifactorial. The discovery of new disease-causing genes and mutations in hereditary forms of retinal distrophies has been the main focus for years. Expertise in human genome analysis and in DNA sequencing technologies derive from the participation - as the only Italian partner - to the Human Genome Project. More recently, we acquired excellent expertise in Next-Generation Sequencing (NGS), particularly in RNA-Sequencing, and we applied this innovative technology to investigate the transcriptome in Down syndrome, type 2 diabetes, and in various types of cancer. The established experience in human genetics, the skills in cellular and molecular biology, coupled with a documented expertise in NGS and RNA-Sequencing, are added values for the ComBOlab.
The research activities are aimed both to support reproducible research initiatives and to fill the gap between rigorous statistical methods and experimental biology. To this purpose, our interests are devoted to the development of novel Bioinformatics tools, such as user-friendly graphical interfaces and databases related to our research activities. Inspired by the reproducible research initiatives, all developed tools are released as open-source to the scientific community or with an open-access format. We are also aimed to implement - as open-source software - all the algorithms that we develop as novel methods. As a complementary part of our activities, we are aimed to setup efficient pipelines combining and using several open-source tools available in literature using the best practice approach. To this purposes, we continuously evaluate and compare different methods either in simulation or on real data, produced in our laboratory. The multidisciplinary composition of the team allows us to evaluate the results both in terms of statistical significance and biological soundness. Current activities include software development for the analysis of epigenomic, transcriptomics data, as well as their integration.
The research activities are motivated by real problems arising in Physics, Medicine and Biology. We have focused on the analysis of high-throughput data by means of statistical and computational methods. Our experience includes Bayesian inference, Denoising, Dimension reduction techniques, Functional data analysis, Multiple hypothesis testing, Multivariate statistics, Non-parametric regression, Supervised and Unsupervised learning, Variable selection and Wavelets. More recently, our interests have been focused on high-dimensional data problems in genomics, transcriptomics and epigenomics. We have recently worked on many problems in statistical genomics, applying methods for the analysis of time-course gene expression data by microarray, for the identification of transcription factor binding sites using variable selection approaches, and methods for NGS data analysis. Current activities include applications to epigenomics and transcriptomics data.