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.
ABSTRACT Blood was collected from a convenience sample of 271 pet cats aged 3 months to 2 years (mean age, 8 months, median and mode, 6 months) between May 1997 and September 1998 in four areas of the United States (southern California, Florida, metropolitan Chicago, and metropolitan Washington, D.C.). Sixty-five (24%) cats had Bartonella henselae bacteremia, and 138 (51%) cats were seropositive for B. henselae . Regional prevalences for bacteremia and seropositivity were highest in Florida (33% and 67%, respectively) and California (28% and 62%, respectively) and lowest in the Washington, D.C. (12% and 28%, respectively) and Chicago (6% and 12%, respectively) areas. No cats bacteremic with B. clarridgeiae were found. The 16S rRNA type was determined for 49 B. henselae isolates. Fourteen of 49 cats (28.6%) were infected with 16S rRNA type I, 32 (65.3%) with 16S rRNA type II, and three (6.1%) were coinfected with 16S rRNA types I and II. Flea infestation was a significant risk factor for B. henselae bacteremia (odds ratio = 2.82, 95% confidence interval, 1.1 to 7.3). Cats ≥13 months old were significantly less likely to be bacteremic than cats ≤6 months old (odds ratio = 0.18, 95% confidence interval, 0.05 to 0.61). Flea infestation, adoption from a shelter or as a stray cat, hunting, and being from Florida or California were significant risk factors for B. henselae seropositivity. DNA fingerprint was significantly associated with region ( P = 0.03) and indoor/outdoor status of cats ( P = 0.03).
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)
