Danilo Cagnina

PhD student

I hold a Bsc. in Biotechnology from the University of Torino and a Msc. in Bioinformatics from the University of Bologna.

During my master’s programme, I have become quite effective at dealing with protein domain and protein structure prediction by implementing Hidden Markov Models, homology modelling and machine-learning methods.

I pursued my master’s thesis at the Centre for Genomic Regulation in Barcelona under the supervision of Professor Cedric Notredame. The aim of the project was to develop an automatic pipeline for evaluating the performances of different programs in producing accurate large-scale protein multiple sequence alignments. The pipeline provides each sequence of the alignment a value of predicted accuracy and a level of uncertainty of this prediction in order to give the user the possibility of selecting the sequences of a large-scale MSA that comply with desired accuracy and uncertainty levels.

I did a five months traineeship at the Joint Research Centre (JRC) (Knowledge for Health & Consumer Safety unit), where my key responsibility consisted of in-silico assessing Genetically Modified (GM) sequence information (NGS and Sanger data) and verifying its compliance with the EU Regulation No 503/2013.

At Martin’s lab, I focus on machine learning methods that help clinicians in fine-tuning tumor diagnosis and prognosis. For instance, I am implementing a tool able to identify the tissue of origin of different cancer types by looking at gene expression profile. This proves useful whenever the true origin of cancer cells is unknown (e.g. cancer of unknown primary, circulating tumor cells, and mislabeled cell lines).