Heterogeneous data integration methods for patient similarity networks.

Briefings in Bioinformatics, 2022, 23(4), 1–26 https://doi.org/10.1093/bib/bbac207 Advance access publication date: 10 June 2022 Review Heterogeneous data integration methods for patient similarity networks Jessica Gliozzo, Marco Mesiti, Marco Notaro, Alessandro Petrini, Alex Patak, Antonio Puertas-Gallardo, Alberto Paccanaro, Giorgio Valentini and Elena Casiraghi Corresponding author: Elena Casiraghi, AnacletoLab, Computer Science Department, Università degli Studi di Milano, Via Celoria 18, 20133 Milan, ITALY. Tel.: +390250316275; Fax: +390250316373; E-mail: Abstract Patient similarity networks (PSNs), where patients are represented as nodes and their similarities as weighted edges, are being increasingly used in clinical research. These networks provide an insightful summary of the relationships among patients and can be exploited by inductive or transductive learning algorithms for the prediction of patient outcome, phenotype and disease risk. PSNs Jessica Gliozzo is a PhD student in computer science enrolled in the Collaborative Doctoral Partnership Program between the University of Milan and the Joint Research Center of European Commission. Her latest research works comprehend the development of a multi-modal semi-supervised method based on patient similarity networks for patients’ outcome prediction; the application of deep neural networks to predict the tissue-specific activity status of cis-regulatory regions in the genome (i.e. promoters and enhancers); the use of compression methods to obtain compact representations of convolution neural networks in the biological domain (e.g. ki67 and TIL-index prediction), showing their advantages when limited computational resources are available. She is author of a few works in the fields of machine learning and bioinformatics. Marco Mesiti is associate professor at the Department of Computer Science Giovanni degli Antoni, Universitá degli Studi di Milano. He has got a master and PhD degree from the University of Genova in 1998 and 2003. His research interest are in the integration, querying and visualization of different kinds of information (structured and semi-structured) according to different data models (relational, graph and nosql). Moreover, he has involved in different projects for protein network integration, protein function prediction and protein networks visualization. On these topics he has published > 100 articles in international conferences and journals. He is associate editor for the Springer Data Science and Engineering Journal and MDPI Applied Sciences. Marco Notaro is a postdoctoral fellow at the Computer Science Department of Milan University. His research interests touch the fields of bioinformatics, computational biology, biological network and machine learning. His main expertise is the analysis and construction of complex biomolecular networks and the design and implementation of output-structured learning algorithms to discover novel gene–disease associations or to predict novel protein function. His PhD paper was awarded by International Medical Informatics Association as one of the best five papers of 2017 in the field of medical informatics. Alessandro Petrini is a postdoctoral researcher at the Department of Computer Science of Universitá degli Studi di Milano. He is currently a member of the Laboratory of Bioinformatics and Computational Biology—AnacletoLab—and his main research is focused on high-performance computing and machine learning. He is author of > 30 articles in international journals and conferences. He designed and developed parallel and accelerated ML algorithms for image and video processing/encoding/compression, omics analisys, graph modeling and analysis, data visualization, neural network compression, MRI volumes processing and analysis. Alex Patak, PhD, MD, graduated in medicine and surgery at the School of Medicine at ‘Universidad Autónoma de Barcelona’, Barcelona (Spain) and holds a Master in medical bioengineering from the ‘Universidad Politécnica de Cataluña’. At the Instituto Municipal de Investigación Médica (Barcelona) he has been working on expert systems for medical diagnostic and did his PhD on computer-assisted medical education after a stage at Dartmouth Medical College in Vermont (USA). Since 1994 works at the Joint Research Centre in Ispra (Italy) where he has been working on three-dimensional medical imaging, and from 2003 to 2017 was responsible for the bioinformatics team at the Molecular Biology and Genomics Unit of the Institute for Health and Consumer Protection in Ispra. He is now a team leader at Knowledge for Health & Consumer Safety and is responsible for the Collaborative Doctoral Partnership Programme in Genomics and Bioinformatics, working on the application of artificial intelligence to omics data and microbiome. Antonio Puertas-Gallardo is an IT project manager at the Joint Research Center (JRC) of the European Commission. He provides high-performance computing (HPC) support to bioinformatics members of the Knowledge for Health and Consumer Safety Unit at JRC, and he has recently begun to collaborate with the unit’s data scientists on natural language processing and machine learning. Alberto Paccanaro is full professor in machine learning and computational biology at the School of Applied Mathematics of the Fundação Getúlio Vargas in Rio de Janeiro and at the Department of Computer Science at Royal Holloway University of London, where he is also Director of the Centre for Systems and Synthetic Biology. He completed his undergraduate studies in computer science at the University of Milan and received his PhD from the University of Toronto in 2002. His research interests are in applying and developing machine learning algorithms for solving problems in molecular biology, medicine and pharmacology and he has led a number of international research projects in this area. Giorgio Valentini is a full professor at the Department of Computer Science, University of Milan (UNIMI). Director for UNIMI of the European doctorate in Genomics and Bioinformatics in collaboration with the Joint Research Center of the European Union. Director of AnacletoLab, Computational Biology and Bioinformatics Laboratory of the Department of Computer Science of the University of Milano. He has been PI in several national and international research projects funded by public and private institutions in the area of bioinformatics, machine learning and big-data analytics. He is author of over 150 scientific publications with peer-review in collaboration with several research groups in Europe and America in the field of bioinformatics, computational biology and machine learning. Elena Casiraghi is associate professor at the Department of Computer Science Giovanni degli Antoni, Universitá degli Studi di Milano. She is co-lead of the AnacletoLab, Computational Biology and Bioinformatics Laboratory of the Department of Computer Science of the University of Milano. Her research (...truncated)