| 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. |
|
|
| 3. |
- Verstappen, S. M. M., et al.
(författare)
-
Methodological Challenges When Comparing Demographic and Clinical Characteristics of International Observational Registries
- 2015
-
Ingår i: Arthritis Care & Research. - : Wiley. - 2151-464X .- 2151-4658. ; 67:12, s. 1637-1645
-
Tidskriftsartikel (refereegranskat)abstract
- Objective. Comparisons of data from different registries can be helpful in understanding variations in many aspects of rheumatoid arthritis (RA). The study aim was to assess and improve the comparability of demographic, clinical, and comorbidity data from 5 international RA registries. Methods. Using predefined definitions, 2 subsets of patients (main cohort and subcohort) from 5 international observational registries (Consortium of Rheumatology Researchers of North America Registry [CORRONA], the Swedish Rheumatology Quality of Care Register [SRR], the Norfolk Arthritis Register [NOAR], the Institute of Rheumatology Rheumatoid Arthritis cohort [IORRA], and CORRONA International) were evaluated and compared. Patients ages >18 years with RA, and present in or recruited to the registry from January 1, 2000, were included in the main cohort. Patients from the main cohort with positive rheumatoid factor and/or erosive RA who had received >= 1 synthetic disease-modifying antirheumatic drug (DMARD), and switched to or added another DMARD, were included in the subcohort at time of treatment switch. Results. Age and sex distributions were fairly similar across the registries. The percentage of patients with a high Disease Activity Score in 28 joints score varied between main cohorts (17.5% IORRA, 18.9% CORRONA, 24.7% NOAR, 27.7% CORRONA International, and 36.8% SRR), with IORRA, CORRONA, and CORRONA International including more prevalent cases of RA; the differences were smaller for the subcohort. Prevalence of comorbidities varied across registries (e.g., coronary artery disease ranged from 1.5% in IORRA to 7.9% in SRR), partly due to the way comorbidity data were captured and general cultural differences; the pattern was similar for the subcohorts. Conclusion. Despite different inclusion criteria for the individual RA registries, it is possible to improve the comparability and interpretability of differences across RA registries by applying well-defined cohort definitions.
|
|
| 4. |
- Agca, R., et al.
(författare)
-
EULAR recommendations for cardiovascular disease risk management in patients with rheumatoid arthritis and other forms of inflammatory joint disorders: 2015/2016 update
- 2017
-
Ingår i: Ann Rheum Dis. - : BMJ. - 0003-4967 .- 1468-2060. ; 76:1, s. 17-28
-
Tidskriftsartikel (refereegranskat)abstract
- Patients with rheumatoid arthritis (RA) and other inflammatory joint disorders (IJD) have increased cardiovascular disease (CVD) risk compared with the general population. In 2009, the European League Against Rheumatism (EULAR) taskforce recommended screening, identification of CVD risk factors and CVD risk management largely based on expert opinion. In view of substantial new evidence, an update was conducted with the aim of producing CVD risk management recommendations for patients with IJD that now incorporates an increasing evidence base. A multidisciplinary steering committee (representing 13 European countries) comprised 26 members including patient representatives, rheumatologists, cardiologists, internists, epidemiologists, a health professional and fellows. Systematic literature searches were performed and evidence was categorised according to standard guidelines. The evidence was discussed and summarised by the experts in the course of a consensus finding and voting process. Three overarching principles were defined. First, there is a higher risk for CVD in patients with RA, and this may also apply to ankylosing spondylitis and psoriatic arthritis. Second, the rheumatologist is responsible for CVD risk management in patients with IJD. Third, the use of non-steroidal anti-inflammatory drugs and corticosteroids should be in accordance with treatment-specific recommendations from EULAR and Assessment of Spondyloarthritis International Society. Ten recommendations were defined, of which one is new and six were changed compared with the 2009 recommendations. Each designated an appropriate evidence support level. The present update extends on the evidence that CVD risk in the whole spectrum of IJD is increased. This underscores the need for CVD risk management in these patients. These recommendations are defined to provide assistance in CVD risk management in IJD, based on expert opinion and scientific evidence.
|
|
| 5. |
|
|
