{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/snect-scalable-network-constrained-tucker","title":"SNeCT: Scalable network constrained Tucker decomposition for integrative multi-platform data analysis","arxiv_id":"1711.08095","date":"2017-11-22","proceeding":null,"authors":["Dongjin Choi","Lee Sael"],"abstract":"Motivation: How do we integratively analyze large-scale multi-platform\ngenomic data that are high dimensional and sparse? Furthermore, how can we\nincorporate prior knowledge, such as the association between genes, in the\nanalysis systematically? Method: To solve this problem, we propose a Scalable\nNetwork Constrained Tucker decomposition method we call SNeCT. SNeCT adopts\nparallel stochastic gradient descent approach on the proposed parallelizable\nnetwork constrained optimization function. SNeCT decomposition is applied to\ntensor constructed from large scale multi-platform multi-cohort cancer data,\nPanCan12, constrained on a network built from PathwayCommons database. Results:\nThe decomposed factor matrices are applied to stratify cancers, to search for\ntop-k similar patients, and to illustrate how the matrices can be used for\npersonalized interpretation. In the stratification test, combined twelve-cohort\ndata is clustered to form thirteen subclasses. The thirteen subclasses have a\nhigh correlation to tissue of origin in addition to other interesting\nobservations, such as clear separation of OV cancers to two groups, and high\nclinical correlation within subclusters formed in cohorts BRCA and UCEC. In the\ntop-k search, a new patient's genomic profile is generated and searched against\nexisting patients based on the factor matrices. The similarity of the top-k\npatient to the query is high for 23 clinical features, including\nestrogen/progesterone receptor statuses of BRCA patients with average precision\nvalue ranges from 0.72 to 0.86 and from 0.68 to 0.86, respectively. We also\nprovide an illustration of how the factor matrices can be used for\ninterpretable personalized analysis of each patient.","url_abs":"http://arxiv.org/abs/1711.08095v2","url_pdf":"http://arxiv.org/pdf/1711.08095v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"snect-scalable-network-constrained-tucker","repo_url":"https://github.com/skywalker5/SNeCT","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}