PMID- 27080919
OWN - NLM
STAT- MEDLINE
DCOM- 20161213
LR  - 20181202
IS  - 1098-2272 (Electronic)
IS  - 0741-0395 (Print)
IS  - 0741-0395 (Linking)
VI  - 40
IP  - 4
DP  - 2016 May
TI  - An Object-Oriented Regression for Building Disease Predictive Models with 
      Multiallelic HLA Genes.
PG  - 315-32
LID - 10.1002/gepi.21968 [doi]
AB  - Recent genome-wide association studies confirm that human leukocyte antigen (HLA) 
      genes have the strongest associations with several autoimmune diseases, including 
      type 1 diabetes (T1D), providing an impetus to reduce this genetic association to 
      practice through an HLA-based disease predictive model. However, conventional 
      model-building methods tend to be suboptimal when predictors are highly 
      polymorphic with many rare alleles combined with complex patterns of sequence 
      homology within and between genes. To circumvent this challenge, we describe an 
      alternative methodology; treating complex genotypes of HLA genes as "objects" or 
      "exemplars," one focuses on systemic associations of disease phenotype with 
      "objects" via similarity measurements. Conceptually, this approach assigns 
      disease risks base on complex genotype profiles instead of specific 
      disease-associated genotypes or alleles. Effectively, it transforms large, 
      discrete, and sparse HLA genotypes into a matrix of similarity-based covariates. 
      By the Kernel representative theorem and machine learning techniques, it uses a 
      penalized likelihood method to select disease-associated exemplars in building 
      predictive models. To illustrate this methodology, we apply it to a T1D study 
      with eight HLA genes (HLA-DRB1, HLA-DRB3, HLA-DRB4, HLA-DRB5, HLA-DQA1, HLA-DQB1, 
      HLA-DPA1, and HLA-DPB1) to build a predictive model. The resulted predictive 
      model has an area under curve of 0.92 in the training set, and 0.89 in the 
      validating set, indicating that this methodology is useful to build predictive 
      models with complex HLA genotypes.
CI  - (c) 2016 WILEY PERIODICALS, INC.
FAU - Zhao, Lue Ping
AU  - Zhao LP
AD  - Division of Public Health Sciences, Fred Hutchinson Cancer Research Center, 
      Seattle, Washington, United States of America.
AD  - Department of Biostatistics, University of Washington School of Public Health, 
      Seattle, Washington, United States of America.
FAU - Bolouri, Hamid
AU  - Bolouri H
AD  - Division of Human Biology, Fred Hutchinson Cancer Research Center, Seattle, 
      Washington, United States of America.
FAU - Zhao, Michael
AU  - Zhao M
AD  - Bellevue High School, Seattle, Washington, United States of America.
FAU - Geraghty, Daniel E
AU  - Geraghty DE
AD  - Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, 
      Washington, United States of America.
FAU - Lernmark, Ake
AU  - Lernmark A
AD  - Department of Clinical Sciences, Lund University/CRC, Skane University Hospital, 
      Malmo, Sweden.
CN  - Better Diabetes Diagnosis Study Group
LA  - eng
GR  - R01 DK026190/DK/NIDDK NIH HHS/United States
GR  - DK26190/DK/NIDDK NIH HHS/United States
GR  - P01 DK053004/DK/NIDDK NIH HHS/United States
GR  - UC4 DK063861/DK/NIDDK NIH HHS/United States
GR  - U01 DK063861/DK/NIDDK NIH HHS/United States
GR  - DK63861/DK/NIDDK NIH HHS/United States
PT  - Journal Article
PT  - Research Support, N.I.H., Extramural
PT  - Research Support, Non-U.S. Gov't
PL  - United States
TA  - Genet Epidemiol
JT  - Genetic epidemiology
JID - 8411723
RN  - 0 (HLA Antigens)
SB  - IM
MH  - *Alleles
MH  - Diabetes Mellitus, Type 1/*genetics
MH  - Genome-Wide Association Study
MH  - Genotype
MH  - HLA Antigens/*genetics
MH  - Humans
MH  - Likelihood Functions
MH  - Linear Models
MH  - *Models, Genetic
MH  - Reproducibility of Results
PMC - PMC4834870
MID - NIHMS766725
OTO - NOTNLM
OT  - generalized linear model
OT  - kernel machine
OT  - multiallelic genotypes
OT  - penalized regression
OT  - prediction
OT  - similarity regression
OT  - statistical learning
COIS- Conflict of Interest Authors declare that there is no conflict of interest with 
      this work.
EDAT- 2016/04/16 06:00
MHDA- 2016/12/15 06:00
PMCR- 2017/05/01
CRDT- 2016/04/16 06:00
PHST- 2015/12/03 00:00 [received]
PHST- 2016/02/11 00:00 [revised]
PHST- 2016/02/17 00:00 [accepted]
PHST- 2016/04/16 06:00 [entrez]
PHST- 2016/04/16 06:00 [pubmed]
PHST- 2016/12/15 06:00 [medline]
PHST- 2017/05/01 00:00 [pmc-release]
AID - 10.1002/gepi.21968 [doi]
PST - ppublish
SO  - Genet Epidemiol. 2016 May;40(4):315-32. doi: 10.1002/gepi.21968.