F Vitali1,2,3, S Marini4, D Pala5, A Demartini5,6, S Montoli5,6, A Zambelli7, R Bellazzi5,6,8. 1. Center for Biomedical Informatics and Biostatistics, The University of Arizona, Tucson, Arizona, USA. 2. BIO5 Institute, The University of Arizona, Tucson, Arizona, USA. 3. Department of Medicine, The University of Arizona, Tucson, AZ, USA. 4. Department of Computational Biology and Bioinformatics, University of Michigan, Ann Arbor, Michigan, USA. 5. Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, PV, Italy. 6. Centre for Health Technologies, University of Pavia, PV, Italy. 7. Oncology Unit, ASST Papa Giovanni XXIII, Bergamo, BG, Italy. 8. IRCCS Istituti Clinici Scientifici Maugeri, Pavia, PV, Italy.
Abstract
OBJECTIVE: Computing patients' similarity is of great interest in precision oncology since it supports clustering and subgroup identification, eventually leading to tailored therapies. The availability of large amounts of biomedical data, characterized by large feature sets and sparse content, motivates the development of new methods to compute patient similarities able to fuse heterogeneous data sources with the available knowledge. MATERIALS AND METHODS: In this work, we developed a data integration approach based on matrix trifactorization to compute patient similarities by integrating several sources of data and knowledge. We assess the accuracy of the proposed method: (1) on several synthetic data sets which similarity structures are affected by increasing levels of noise and data sparsity, and (2) on a real data set coming from an acute myeloid leukemia (AML) study. The results obtained are finally compared with the ones of traditional similarity calculation methods. RESULTS: In the analysis of the synthetic data set, where the ground truth is known, we measured the capability of reconstructing the correct clusters, while in the AML study we evaluated the Kaplan-Meier curves obtained with the different clusters and measured their statistical difference by means of the log-rank test. In presence of noise and sparse data, our data integration method outperform other techniques, both in the synthetic and in the AML data. DISCUSSION: In case of multiple heterogeneous data sources, a matrix trifactorization technique can successfully fuse all the information in a joint model. We demonstrated how this approach can be efficiently applied to discover meaningful patient similarities and therefore may be considered a reliable data driven strategy for the definition of new research hypothesis for precision oncology. CONCLUSION: The better performance of the proposed approach presents an advantage over previous methods to provide accurate patient similarities supporting precision medicine.
OBJECTIVE: Computing patients' similarity is of great interest in precision oncology since it supports clustering and subgroup identification, eventually leading to tailored therapies. The availability of large amounts of biomedical data, characterized by large feature sets and sparse content, motivates the development of new methods to compute patient similarities able to fuse heterogeneous data sources with the available knowledge. MATERIALS AND METHODS: In this work, we developed a data integration approach based on matrix trifactorization to compute patient similarities by integrating several sources of data and knowledge. We assess the accuracy of the proposed method: (1) on several synthetic data sets which similarity structures are affected by increasing levels of noise and data sparsity, and (2) on a real data set coming from an acute myeloid leukemia (AML) study. The results obtained are finally compared with the ones of traditional similarity calculation methods. RESULTS: In the analysis of the synthetic data set, where the ground truth is known, we measured the capability of reconstructing the correct clusters, while in the AML study we evaluated the Kaplan-Meier curves obtained with the different clusters and measured their statistical difference by means of the log-rank test. In presence of noise and sparse data, our data integration method outperform other techniques, both in the synthetic and in the AML data. DISCUSSION: In case of multiple heterogeneous data sources, a matrix trifactorization technique can successfully fuse all the information in a joint model. We demonstrated how this approach can be efficiently applied to discover meaningful patient similarities and therefore may be considered a reliable data driven strategy for the definition of new research hypothesis for precision oncology. CONCLUSION: The better performance of the proposed approach presents an advantage over previous methods to provide accurate patient similarities supporting precision medicine.
Authors: Richard F Schlenk; Konstanze Döhner; Jürgen Krauter; Stefan Fröhling; Andrea Corbacioglu; Lars Bullinger; Marianne Habdank; Daniela Späth; Michael Morgan; Axel Benner; Brigitte Schlegelberger; Gerhard Heil; Arnold Ganser; Hartmut Döhner Journal: N Engl J Med Date: 2008-05-01 Impact factor: 91.245
Authors: Thomas J Hudson; Warwick Anderson; Axel Artez; Anna D Barker; Cindy Bell; Rosa R Bernabé; M K Bhan; Fabien Calvo; Iiro Eerola; Daniela S Gerhard; Alan Guttmacher; Mark Guyer; Fiona M Hemsley; Jennifer L Jennings; David Kerr; Peter Klatt; Patrik Kolar; Jun Kusada; David P Lane; Frank Laplace; Lu Youyong; Gerd Nettekoven; Brad Ozenberger; Jane Peterson; T S Rao; Jacques Remacle; Alan J Schafer; Tatsuhiro Shibata; Michael R Stratton; Joseph G Vockley; Koichi Watanabe; Huanming Yang; Matthew M F Yuen; Bartha M Knoppers; Martin Bobrow; Anne Cambon-Thomsen; Lynn G Dressler; Stephanie O M Dyke; Yann Joly; Kazuto Kato; Karen L Kennedy; Pilar Nicolás; Michael J Parker; Emmanuelle Rial-Sebbag; Carlos M Romeo-Casabona; Kenna M Shaw; Susan Wallace; Georgia L Wiesner; Nikolajs Zeps; Peter Lichter; Andrew V Biankin; Christian Chabannon; Lynda Chin; Bruno Clément; Enrique de Alava; Françoise Degos; Martin L Ferguson; Peter Geary; D Neil Hayes; Thomas J Hudson; Amber L Johns; Arek Kasprzyk; Hidewaki Nakagawa; Robert Penny; Miguel A Piris; Rajiv Sarin; Aldo Scarpa; Tatsuhiro Shibata; Marc van de Vijver; P Andrew Futreal; Hiroyuki Aburatani; Mónica Bayés; David D L Botwell; Peter J Campbell; Xavier Estivill; Daniela S Gerhard; Sean M Grimmond; Ivo Gut; Martin Hirst; Carlos López-Otín; Partha Majumder; Marco Marra; John D McPherson; Hidewaki Nakagawa; Zemin Ning; Xose S Puente; Yijun Ruan; Tatsuhiro Shibata; Michael R Stratton; Hendrik G Stunnenberg; Harold Swerdlow; Victor E Velculescu; Richard K Wilson; Hong H Xue; Liu Yang; Paul T Spellman; Gary D Bader; Paul C Boutros; Peter J Campbell; Paul Flicek; Gad Getz; Roderic Guigó; Guangwu Guo; David Haussler; Simon Heath; Tim J Hubbard; Tao Jiang; Steven M Jones; Qibin Li; Nuria López-Bigas; Ruibang Luo; Lakshmi Muthuswamy; B F Francis Ouellette; John V Pearson; Xose S Puente; Victor Quesada; Benjamin J Raphael; Chris Sander; Tatsuhiro Shibata; Terence P Speed; Lincoln D Stein; Joshua M Stuart; Jon W Teague; Yasushi Totoki; Tatsuhiko Tsunoda; Alfonso Valencia; David A Wheeler; Honglong Wu; Shancen Zhao; Guangyu Zhou; Lincoln D Stein; Roderic Guigó; Tim J Hubbard; Yann Joly; Steven M Jones; Arek Kasprzyk; Mark Lathrop; Nuria López-Bigas; B F Francis Ouellette; Paul T Spellman; Jon W Teague; Gilles Thomas; Alfonso Valencia; Teruhiko Yoshida; Karen L Kennedy; Myles Axton; Stephanie O M Dyke; P Andrew Futreal; Daniela S Gerhard; Chris Gunter; Mark Guyer; Thomas J Hudson; John D McPherson; Linda J Miller; Brad Ozenberger; Kenna M Shaw; Arek Kasprzyk; Lincoln D Stein; Junjun Zhang; Syed A Haider; Jianxin Wang; Christina K Yung; Anthony Cros; Anthony Cross; Yong Liang; Saravanamuttu Gnaneshan; Jonathan Guberman; Jack Hsu; Martin Bobrow; Don R C Chalmers; Karl W Hasel; Yann Joly; Terry S H Kaan; Karen L Kennedy; Bartha M Knoppers; William W Lowrance; Tohru Masui; Pilar Nicolás; Emmanuelle Rial-Sebbag; Laura Lyman Rodriguez; Catherine Vergely; Teruhiko Yoshida; Sean M Grimmond; Andrew V Biankin; David D L Bowtell; Nicole Cloonan; Anna deFazio; James R Eshleman; Dariush Etemadmoghadam; Brooke B Gardiner; Brooke A Gardiner; James G Kench; Aldo Scarpa; Robert L Sutherland; Margaret A Tempero; Nicola J Waddell; Peter J Wilson; John D McPherson; Steve Gallinger; Ming-Sound Tsao; Patricia A Shaw; Gloria M Petersen; Debabrata Mukhopadhyay; Lynda Chin; Ronald A DePinho; Sarah Thayer; Lakshmi Muthuswamy; Kamran Shazand; Timothy Beck; Michelle Sam; Lee Timms; Vanessa Ballin; Youyong Lu; Jiafu Ji; Xiuqing Zhang; Feng Chen; Xueda Hu; Guangyu Zhou; Qi Yang; Geng Tian; Lianhai Zhang; Xiaofang Xing; Xianghong Li; Zhenggang Zhu; Yingyan Yu; Jun Yu; Huanming Yang; Mark Lathrop; Jörg Tost; Paul Brennan; Ivana Holcatova; David Zaridze; Alvis Brazma; Lars Egevard; Egor Prokhortchouk; Rosamonde Elizabeth Banks; Mathias Uhlén; Anne Cambon-Thomsen; Juris Viksna; Fredrik Ponten; Konstantin Skryabin; Michael R Stratton; P Andrew Futreal; Ewan Birney; Ake Borg; Anne-Lise Børresen-Dale; Carlos Caldas; John A Foekens; Sancha Martin; Jorge S Reis-Filho; Andrea L Richardson; Christos Sotiriou; Hendrik G Stunnenberg; Giles Thoms; Marc van de Vijver; Laura van't Veer; Fabien Calvo; Daniel Birnbaum; Hélène Blanche; Pascal Boucher; Sandrine Boyault; Christian Chabannon; Ivo Gut; Jocelyne D Masson-Jacquemier; Mark Lathrop; Iris Pauporté; Xavier Pivot; Anne Vincent-Salomon; Eric Tabone; Charles Theillet; Gilles Thomas; Jörg Tost; Isabelle Treilleux; Fabien Calvo; Paulette Bioulac-Sage; Bruno Clément; Thomas Decaens; Françoise Degos; Dominique Franco; Ivo Gut; Marta Gut; Simon Heath; Mark Lathrop; Didier Samuel; Gilles Thomas; Jessica Zucman-Rossi; Peter Lichter; Roland Eils; Benedikt Brors; Jan O Korbel; Andrey Korshunov; Pablo Landgraf; Hans Lehrach; Stefan Pfister; Bernhard Radlwimmer; Guido Reifenberger; Michael D Taylor; Christof von Kalle; Partha P Majumder; Rajiv Sarin; T S Rao; M K Bhan; Aldo Scarpa; Paolo Pederzoli; Rita A Lawlor; Massimo Delledonne; Alberto Bardelli; Andrew V Biankin; Sean M Grimmond; Thomas Gress; David Klimstra; Giuseppe Zamboni; Tatsuhiro Shibata; Yusuke Nakamura; Hidewaki Nakagawa; Jun Kusada; Tatsuhiko Tsunoda; Satoru Miyano; Hiroyuki Aburatani; Kazuto Kato; Akihiro Fujimoto; Teruhiko Yoshida; Elias Campo; Carlos López-Otín; Xavier Estivill; Roderic Guigó; Silvia de Sanjosé; Miguel A Piris; Emili Montserrat; Marcos González-Díaz; Xose S Puente; Pedro Jares; Alfonso Valencia; Heinz Himmelbauer; Heinz Himmelbaue; Victor Quesada; Silvia Bea; Michael R Stratton; P Andrew Futreal; Peter J Campbell; Anne Vincent-Salomon; Andrea L Richardson; Jorge S Reis-Filho; Marc van de Vijver; Gilles Thomas; Jocelyne D Masson-Jacquemier; Samuel Aparicio; Ake Borg; Anne-Lise Børresen-Dale; Carlos Caldas; John A Foekens; Hendrik G Stunnenberg; Laura van't Veer; Douglas F Easton; Paul T Spellman; Sancha Martin; Anna D Barker; Lynda Chin; Francis S Collins; Carolyn C Compton; Martin L Ferguson; Daniela S Gerhard; Gad Getz; Chris Gunter; Alan Guttmacher; Mark Guyer; D Neil Hayes; Eric S Lander; Brad Ozenberger; Robert Penny; Jane Peterson; Chris Sander; Kenna M Shaw; Terence P Speed; Paul T Spellman; Joseph G Vockley; David A Wheeler; Richard K Wilson; Thomas J Hudson; Lynda Chin; Bartha M Knoppers; Eric S Lander; Peter Lichter; Lincoln D Stein; Michael R Stratton; Warwick Anderson; Anna D Barker; Cindy Bell; Martin Bobrow; Wylie Burke; Francis S Collins; Carolyn C Compton; Ronald A DePinho; Douglas F Easton; P Andrew Futreal; Daniela S Gerhard; Anthony R Green; Mark Guyer; Stanley R Hamilton; Tim J Hubbard; Olli P Kallioniemi; Karen L Kennedy; Timothy J Ley; Edison T Liu; Youyong Lu; Partha Majumder; Marco Marra; Brad Ozenberger; Jane Peterson; Alan J Schafer; Paul T Spellman; Hendrik G Stunnenberg; Brandon J Wainwright; Richard K Wilson; Huanming Yang Journal: Nature Date: 2010-04-15 Impact factor: 49.962
Authors: Andrew Chatr-Aryamontri; Bobby-Joe Breitkreutz; Rose Oughtred; Lorrie Boucher; Sven Heinicke; Daici Chen; Chris Stark; Ashton Breitkreutz; Nadine Kolas; Lara O'Donnell; Teresa Reguly; Julie Nixon; Lindsay Ramage; Andrew Winter; Adnane Sellam; Christie Chang; Jodi Hirschman; Chandra Theesfeld; Jennifer Rust; Michael S Livstone; Kara Dolinski; Mike Tyers Journal: Nucleic Acids Res Date: 2014-11-26 Impact factor: 19.160
Authors: Joanna S Amberger; Carol A Bocchini; François Schiettecatte; Alan F Scott; Ada Hamosh Journal: Nucleic Acids Res Date: 2014-11-26 Impact factor: 19.160
Authors: Jessica Gliozzo; Marco Mesiti; Marco Notaro; Alessandro Petrini; Alex Patak; Antonio Puertas-Gallardo; Alberto Paccanaro; Giorgio Valentini; Elena Casiraghi Journal: Brief Bioinform Date: 2022-07-18 Impact factor: 13.994