Md Jamiul Jahid1, Tim H Huang2, Jianhua Ruan2. 1. Department of Computer Science, University of Texas at San Antonio, San Antonio, TX 78249, USA, Department of Molecular Medicine and Cancer Therapy & Research Center, University of Texas Health Science Center, San Antonio, TX 78229, USA. 2. Department of Computer Science, University of Texas at San Antonio, San Antonio, TX 78249, USA, Department of Molecular Medicine and Cancer Therapy & Research Center, University of Texas Health Science Center, San Antonio, TX 78229, USADepartment of Computer Science, University of Texas at San Antonio, San Antonio, TX 78249, USA, Department of Molecular Medicine and Cancer Therapy & Research Center, University of Texas Health Science Center, San Antonio, TX 78229, USA.
Abstract
MOTIVATION: Metastasis prediction is a well-known problem in breast cancer research. As breast cancer is a complex and heterogeneous disease with many molecular subtypes, predictive models trained for one cohort often perform poorly on other cohorts, and a combined model may be suboptimal for individual patients. Furthermore, attempting to develop subtype-specific models is hindered by the ambiguity and stereotypical definitions of subtypes. RESULTS: Here, we propose a personalized approach by relaxing the definition of breast cancer subtypes. We assume that each patient belongs to a distinct subtype, defined implicitly by a set of patients with similar molecular characteristics, and construct a different predictive model for each patient, using as training data, only the patients defining the subtype. To increase robustness, we also develop a committee-based prediction method by pooling together multiple personalized models. Using both intra- and inter-dataset validations, we show that our approach can significantly improve the prediction accuracy of breast cancer metastasis compared with several popular approaches, especially on those hard-to-learn cases. Furthermore, we find that breast cancer patients belonging to different canonical subtypes tend to have different predictive models and gene signatures, suggesting that metastasis in different canonical subtypes are likely governed by different molecular mechanisms. AVAILABILITY AND IMPLEMENTATION: Source code implemented in MATLAB and Java available at www.cs.utsa.edu/∼jruan/PCC/.
MOTIVATION: Metastasis prediction is a well-known problem in breast cancer research. As breast cancer is a complex and heterogeneous disease with many molecular subtypes, predictive models trained for one cohort often perform poorly on other cohorts, and a combined model may be suboptimal for individual patients. Furthermore, attempting to develop subtype-specific models is hindered by the ambiguity and stereotypical definitions of subtypes. RESULTS: Here, we propose a personalized approach by relaxing the definition of breast cancer subtypes. We assume that each patient belongs to a distinct subtype, defined implicitly by a set of patients with similar molecular characteristics, and construct a different predictive model for each patient, using as training data, only the patients defining the subtype. To increase robustness, we also develop a committee-based prediction method by pooling together multiple personalized models. Using both intra- and inter-dataset validations, we show that our approach can significantly improve the prediction accuracy of breast cancer metastasis compared with several popular approaches, especially on those hard-to-learn cases. Furthermore, we find that breast cancerpatients belonging to different canonical subtypes tend to have different predictive models and gene signatures, suggesting that metastasis in different canonical subtypes are likely governed by different molecular mechanisms. AVAILABILITY AND IMPLEMENTATION: Source code implemented in MATLAB and Java available at www.cs.utsa.edu/∼jruan/PCC/.
Authors: Laura M Heiser; Anguraj Sadanandam; Wen-Lin Kuo; Stephen C Benz; Theodore C Goldstein; Sam Ng; William J Gibb; Nicholas J Wang; Safiyyah Ziyad; Frances Tong; Nora Bayani; Zhi Hu; Jessica I Billig; Andrea Dueregger; Sophia Lewis; Lakshmi Jakkula; James E Korkola; Steffen Durinck; François Pepin; Yinghui Guan; Elizabeth Purdom; Pierre Neuvial; Henrik Bengtsson; Kenneth W Wood; Peter G Smith; Lyubomir T Vassilev; Bryan T Hennessy; Joel Greshock; Kurtis E Bachman; Mary Ann Hardwicke; John W Park; Laurence J Marton; Denise M Wolf; Eric A Collisson; Richard M Neve; Gordon B Mills; Terence P Speed; Heidi S Feiler; Richard F Wooster; David Haussler; Joshua M Stuart; Joe W Gray; Paul T Spellman Journal: Proc Natl Acad Sci U S A Date: 2011-10-14 Impact factor: 11.205
Authors: Laura J van 't Veer; Hongyue Dai; Marc J van de Vijver; Yudong D He; Augustinus A M Hart; Mao Mao; Hans L Peterse; Karin van der Kooy; Matthew J Marton; Anke T Witteveen; George J Schreiber; Ron M Kerkhoven; Chris Roberts; Peter S Linsley; René Bernards; Stephen H Friend Journal: Nature Date: 2002-01-31 Impact factor: 49.962
Authors: T Sørlie; C M Perou; R Tibshirani; T Aas; S Geisler; H Johnsen; T Hastie; M B Eisen; M van de Rijn; S S Jeffrey; T Thorsen; H Quist; J C Matese; P O Brown; D Botstein; P E Lønning; A L Børresen-Dale Journal: Proc Natl Acad Sci U S A Date: 2001-09-11 Impact factor: 11.205
Authors: Takasuke Yamachika; J Lawrence Werther; Carol Bodian; Mark Babyatsky; Masae Tatematsu; Yoshitaka Yamamura; Anli Chen; Steven Itzkowitz Journal: Clin Cancer Res Date: 2002-05 Impact factor: 12.531
Authors: Marc J van de Vijver; Yudong D He; Laura J van't Veer; Hongyue Dai; Augustinus A M Hart; Dorien W Voskuil; George J Schreiber; Johannes L Peterse; Chris Roberts; Matthew J Marton; Mark Parrish; Douwe Atsma; Anke Witteveen; Annuska Glas; Leonie Delahaye; Tony van der Velde; Harry Bartelink; Sjoerd Rodenhuis; Emiel T Rutgers; Stephen H Friend; René Bernards Journal: N Engl J Med Date: 2002-12-19 Impact factor: 91.245
Authors: Bisrat G Debeb; Evan N Cohen; Kimberly Boley; Erik M Freiter; Li Li; Fredika M Robertson; James M Reuben; Massimo Cristofanilli; Thomas A Buchholz; Wendy A Woodward Journal: Breast Cancer Res Treat Date: 2012-05-01 Impact factor: 4.872
Authors: Therese Sorlie; Robert Tibshirani; Joel Parker; Trevor Hastie; J S Marron; Andrew Nobel; Shibing Deng; Hilde Johnsen; Robert Pesich; Stephanie Geisler; Janos Demeter; Charles M Perou; Per E Lønning; Patrick O Brown; Anne-Lise Børresen-Dale; David Botstein Journal: Proc Natl Acad Sci U S A Date: 2003-06-26 Impact factor: 12.779
Authors: Mattia C F Prosperi; Sarah L Ingham; Anthony Howell; Fiona Lalloo; Iain E Buchan; Dafydd Gareth Evans Journal: BMC Med Inform Decis Mak Date: 2014-10-01 Impact factor: 2.796
Authors: Maryam Farhadian; Hossein Mahjub; Jalal Poorolajal; Abbas Moghimbeigi; Muharram Mansoorizadeh Journal: Osong Public Health Res Perspect Date: 2014-11-01