1
|
Liao Q, Liu Q, Razak FA. Hypergraph regularized nonnegative triple decomposition for multiway data analysis. Sci Rep 2024; 14:9098. [PMID: 38643209 PMCID: PMC11032410 DOI: 10.1038/s41598-024-59300-3] [Citation(s) in RCA: 0] [Impact Index Per Article: 0] [Reference Citation Analysis] [Abstract] [Key Words] [Grants] [Track Full Text] [Journal Information] [Subscribe] [Scholar Register] [Received: 01/09/2024] [Accepted: 04/09/2024] [Indexed: 04/22/2024] Open
Abstract
Tucker decomposition is widely used for image representation, data reconstruction, and machine learning tasks, but the calculation cost for updating the Tucker core is high. Bilevel form of triple decomposition (TriD) overcomes this issue by decomposing the Tucker core into three low-dimensional third-order factor tensors and plays an important role in the dimension reduction of data representation. TriD, on the other hand, is incapable of precisely encoding similarity relationships for tensor data with a complex manifold structure. To address this shortcoming, we take advantage of hypergraph learning and propose a novel hypergraph regularized nonnegative triple decomposition for multiway data analysis that employs the hypergraph to model the complex relationships among the raw data. Furthermore, we develop a multiplicative update algorithm to solve our optimization problem and theoretically prove its convergence. Finally, we perform extensive numerical tests on six real-world datasets, and the results show that our proposed algorithm outperforms some state-of-the-art methods.
Collapse
Affiliation(s)
- Qingshui Liao
- Department of Mathematical Sciences, Faculty of Science & Technology, Universiti Kebangsaan Malaysia, 43600, Bangi, Selangor, Malaysia.
- School of Mathematical Sciences, Guizhou Normal University, Guiyang, 550025, People's Republic of China.
| | - Qilong Liu
- School of Mathematical Sciences, Guizhou Normal University, Guiyang, 550025, People's Republic of China
| | - Fatimah Abdul Razak
- Department of Mathematical Sciences, Faculty of Science & Technology, Universiti Kebangsaan Malaysia, 43600, Bangi, Selangor, Malaysia.
| |
Collapse
|
2
|
Balasubramanyam A, Redondo MJ, Craigen W, Dai H, Davis A, Desai D, Dussan M, Faruqi J, Gaba R, Gonzalez I, Jhangiani S, Kubota-Mishra E, Liu P, Murdock D, Posey J, Ram N, Sabo A, Sisley S, Tosur M, Venner E, Astudillo M, Cardenas A, Fang MA, Hattery E, Ideouzu A, Jimenez J, Kikani N, Montes G, O’Brien NG, Wong LJ, Goland R, Chung WK, Evans A, Gandica R, Leibel R, Mofford K, Pring J, Evans-Molina C, Anwar F, Monaco G, Neyman A, Saeed Z, Sims E, Spall M, Hernandez-Perez M, Mather K, Moors K, Udler MS, Florez JC, Calverley M, Chen V, Chu K, Cromer S, Deutsch A, Faciebene M, Greaux E, Koren D, Kreienkamp R, Larkin M, Marshall W, Ricevuto P, Sabean A, Thangthaeng N, Han C, Sherwood J, Billings LK, Banerji MA, Bally K, Brown N, Ji B, Soni L, Lee M, Abrams J, Thomas L, Abrams J, Skiwiersky S, Philipson LH, Greeley SAW, Bell G, Banogon S, Desai J, Ehrmann D, Letourneau-Freiberg LR, Naylor RN, Papciak E, Friedman Ross L, Sundaresan M, Bender C, Tian P, Rasouli N, Kashkouli MB, Baker C, Her A, King C, Pyreddy A, Singh V, Barklow J, Farhat N, Lorch R, Odean C, Schleis G, Underkofler C, Pollin TI, Bryan H, Maloney K, Miller R, Newton P, Nikita ME, Nwaba D, Silver K, Tiner J, Whitlatch H, Palmer K, Riley S, Streeten E, Oral EA, Broome D, Dill Gomes A, Foss de Freitas M, Gregg B, Grigoryan S, Imam S, Sonmez Ince M, Neidert A, Richison C, Akinci B, Hench R, Buse J, Armstrong C, Christensen C, Diner J, Fraser R, Fulghum K, Ghorbani T, Kass A, Klein K, Kirkman MS, Hirsch IB, Baran J, Dong X, Kahn SE, Khakpour D, Mandava P, Sameshima L, Kalerus T, Pihoker C, Loots B, Santarelli K, Pascual C, Niswender K, Edwards N, Gregory J, Powers A, Ramirez A, Scott J, Smith J, Urano F, Hughes J, Hurst S, McGill J, Stone S, May J, Krischer JP, Adusumalli R, Albritton B, Aquino A, Bransford P, Cadigan N, Gandolfo L, Garmeson J, Gomes J, Gowing R, Karges C, Kirk C, Muller S, Morissette J, Parikh HM, Perez-Laras F, Remedios CL, Ruiz P, Sulman N, Toth M, Wurmser L, Eberhard C, Fiske S, Hutchinson B, Nekkanti S, Wood R, Florez JC, Alkanaq A, Brandes M, Burtt N, Flannick J, Olorunfemi P, Udler MS, Caulkins L, Wasserfall C, Winter W, Pittman D, Akolkar B, Lee C, Carey DJ, Hood D, Marcovina SM, Newgard CB. The Rare and Atypical Diabetes Network (RADIANT) Study: Design and Early Results. Diabetes Care 2023; 46:1265-1270. [PMID: 37104866 PMCID: PMC10234756 DOI: 10.2337/dc22-2440] [Citation(s) in RCA: 2] [Impact Index Per Article: 2.0] [Reference Citation Analysis] [Abstract] [MESH Headings] [Grants] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 12/15/2022] [Accepted: 03/27/2023] [Indexed: 04/29/2023]
Abstract
OBJECTIVE The Rare and Atypical Diabetes Network (RADIANT) will perform a study of individuals and, if deemed informative, a study of their family members with uncharacterized forms of diabetes. RESEARCH DESIGN AND METHODS The protocol includes genomic (whole-genome [WGS], RNA, and mitochondrial sequencing), phenotypic (vital signs, biometric measurements, questionnaires, and photography), metabolomics, and metabolic assessments. RESULTS Among 122 with WGS results of 878 enrolled individuals, a likely pathogenic variant in a known diabetes monogenic gene was found in 3 (2.5%), and six new monogenic variants have been identified in the SMAD5, PTPMT1, INS, NFKB1, IGF1R, and PAX6 genes. Frequent phenotypic clusters are lean type 2 diabetes, autoantibody-negative and insulin-deficient diabetes, lipodystrophic diabetes, and new forms of possible monogenic or oligogenic diabetes. CONCLUSIONS The analyses will lead to improved means of atypical diabetes identification. Genetic sequencing can identify new variants, and metabolomics and transcriptomics analysis can identify novel mechanisms and biomarkers for atypical disease.
Collapse
|