| 1. |
- Romagnoni, A, et al.
(författare)
-
Comparative performances of machine learning methods for classifying Crohn Disease patients using genome-wide genotyping data
- 2019
-
Ingår i: Scientific reports. - : Springer Science and Business Media LLC. - 2045-2322. ; 9:1, s. 10351-
-
Tidskriftsartikel (refereegranskat)abstract
- Crohn Disease (CD) is a complex genetic disorder for which more than 140 genes have been identified using genome wide association studies (GWAS). However, the genetic architecture of the trait remains largely unknown. The recent development of machine learning (ML) approaches incited us to apply them to classify healthy and diseased people according to their genomic information. The Immunochip dataset containing 18,227 CD patients and 34,050 healthy controls enrolled and genotyped by the international Inflammatory Bowel Disease genetic consortium (IIBDGC) has been re-analyzed using a set of ML methods: penalized logistic regression (LR), gradient boosted trees (GBT) and artificial neural networks (NN). The main score used to compare the methods was the Area Under the ROC Curve (AUC) statistics. The impact of quality control (QC), imputing and coding methods on LR results showed that QC methods and imputation of missing genotypes may artificially increase the scores. At the opposite, neither the patient/control ratio nor marker preselection or coding strategies significantly affected the results. LR methods, including Lasso, Ridge and ElasticNet provided similar results with a maximum AUC of 0.80. GBT methods like XGBoost, LightGBM and CatBoost, together with dense NN with one or more hidden layers, provided similar AUC values, suggesting limited epistatic effects in the genetic architecture of the trait. ML methods detected near all the genetic variants previously identified by GWAS among the best predictors plus additional predictors with lower effects. The robustness and complementarity of the different methods are also studied. Compared to LR, non-linear models such as GBT or NN may provide robust complementary approaches to identify and classify genetic markers.
|
|
| 2. |
- Xu, Jianfeng, et al.
(författare)
-
Estimation of absolute risk for prostate cancer using genetic markers and family history
- 2009
-
Ingår i: The Prostate. - : Wiley. - 0270-4137 .- 1097-0045. ; 69:14, s. 1565-1572
-
Tidskriftsartikel (refereegranskat)abstract
- BACKGROUND: Multiple DNA sequence variants in the form of single-nucleotide polymorphisms (SNPs) have been found to be reproducibly associated with prostate cancer (PCa) risk. METHODS: Absolute risk for PCa among men with various numbers of inherited risk alleles and family history of PCa was estimated in a population-based case-control study in Sweden (2,893 cases and 1,781 controls), and a nested case-control study from the Prostate, Lung, Colon and Ovarian (PLCO) Cancer Screening Trial in the U.S. (1,172 cases and 1,157 controls). RESULTS: Increased number of risk alleles and positive family history were independently associated with PCa risk. Considering men with 11 risk alleles (mode) and negative family history as having baseline risk, men who had >or=14 risk alleles and positive family history had an odds ratio (OR) of 4.92 [95% confidence interval (CI): 3.64-6.64] in the Swedish study. These associations were confirmed in the U.S. population. Once a man's SNP genotypes and family history are known, his absolute risk for PCa can be readily calculated and easily interpreted. For example, 55-year-old men with a family history and >or=14 risk alleles have a 52% and 41% risk of being diagnosed with PCa in the next 20 years in the Swedish and U.S. populations, respectively. In comparison, without knowledge of genotype and family history, these men had an average population absolute risk of 13%. CONCLUSION: This risk prediction model may be used to identify men at considerably elevated PCa risk who may be selected for chemoprevention.
|
|
