ComBat normalization markedly increases cross-study prediction when the training and test data have different feature distribution with XGBoost classifiers. حسب Yilin Gao (4335571)

The Transition from a Compensatory Increase to a Decrease in C-peptide During the Progression to Type 1 Diabetes and Its Relation to Risk حسب Heba M Ismail (9568748)

"…This change was mostly attributable to change in the late AUC C-peptide response (i.e., 60-to-120-minute AUC C-peptide). Median Index60 values of  DPT-1 (1.44) and TNPTP (1.05) progressors to type 1 diabetes 1.5 years pre-diagnosis (time of transition from increasing to decreasing AUC C-peptide) were used as thresholds to identify individuals at high risk for type 1 diabetes in the full cohort at baseline (5-year risk of 0.75-0.88 for those above thresholds). …"

The ROC curves for the four serum proteins with the highest AUC values. حسب Matilda Holm (5076659)

<i>E</i> values weakly decrease with absolute latitude. حسب Dimitrios—Georgios Kontopoulos (9530566)

Top features with an increased risk of overdose (Logistic regression). حسب David Warren (3408077)

ComBat normalization markedly increases cross-study prediction when the studies have batch effects when using LASSO classifiers. حسب Yilin Gao (4335571)

ComBat normalization markedly increases cross-study prediction when the studies have batch effects when using XGBoost classifiers. حسب Yilin Gao (4335571)

The AUC values of different sampling methods in the cumulative sample size. حسب Xi Li (19208)

Decreasing genome breadth of coverage in samples with high Cq values. حسب Paul Y. Kim (7491026)

ComBat normalization increases cross-study prediction when the training and test data have different feature distribution with LASSO classifiers. حسب Yilin Gao (4335571)

ComBat normalization markedly increases cross-study prediction when the training and test data have varying degrees of overlap in disease-associated OTUs. حسب Yilin Gao (4335571)

Box plots showing the normalized abundances of the four serum proteins with the highest AUC values in non-PSC controls and PSC patients. حسب Matilda Holm (5076659)

ComBat normalization markedly increases cross-study prediction when the training and test data have different disease models when using XGBoost classifiers. حسب Yilin Gao (4335571)

ComBat normalization markedly increases cross-study prediction when the training and test data have different disease models when using LASSO classifiers. حسب Yilin Gao (4335571)

ComBat normalization markedly increases cross-study prediction when the training and test data have different disease models when using SVM classifiers. حسب Yilin Gao (4335571)

Decreasing pixel differences caused by worm movement correspond to decreasing locomotive health. حسب Areta Jushaj (8525655)

Effect of parameter <i>a</i> on the model equations by increasing and decreasing value of <i>a</i> to 25%. حسب Shakoor Ahmad (168711)

"…<p><b>(a)</b> By increasing the value ‘<i>a</i>’ from 0.107 to 0.134. <b>(b)</b> By decreasing the value ‘<i>a</i>’ from 0.107 to 0.08.…"

TITAN decreaser macroinvertebrate heatmap Heatmap of sensitive (i.e., <i>z-</i>) macroinvertebrate taxa (y-axis) to at least one of the five stressors, in decreasing order of numbe... حسب Brent J. Bellinger (21156150)

Top features with an increased risk of overdose after excluding socio-economic features (Logistic regression). حسب David Warren (3408077)

Species occurrence along the NO₃¯ gradient (sorted by decreasing 25th percentile values). حسب Willem Kaijser (12872371)