2026-10-14T15:18:45Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33816697 2026-09-15T18:20:14Z category_13 category_21 category_24 category_734 category_811 category_64 category_132 portal_5 item_type_3 month_year_09_2026

<p>N-feature model performance metrics. Classification accuracy metrics for models trained on the top three, five, seven, nine, and twelve most important features. Classification metrics of training and testing specimens are shown in Train columns and Test columns. Bootstrapping was used to calculate average values and 95% confidence interval, shown in each cell.</p> Jacob H. Elnaggar (14853346) John W. Lammons (17620176) Caleb M. Ardizzone (10583735) Kristal J. Aaron (12047007) Clayton Jacobs (24950511) Keonte J. Graves (19370020) Sheridan D. George (24950514) Megan Amerson-Brown (24950517) Meng Luo (196199) Ashutosh Tamhane (9951397) Paweł Łaniewski (624510) Alison J. Quayle (6669437) Melissa M. Herbst-Kralovetz (11871161) Nuno Cerca (163297) Christina A. Muzny (10271067) Christopher M. Taylor (7588550) Genetics Biotechnology Evolutionary Biology Biological Sciences not elsewhere classified Mathematical Sciences not elsewhere classified Cancer Infectious Diseases xlink "> anns vaginal dysbiosis associated vaginal bacterial taxa five taxa (< level test set shap analysis identified ann achieved 93 95 %, specificity identify taxa associated lactobacillus crispatus </ validation produced lower external validation indicates 826 ± 0 shap analysis strongly associated lactobacillus </ ± 8 lower performance identify women external validation level cross mulieris </ megamonas </ iners </ gardnerella </ variable performance symptom onset relative abundance model predictions model performance infectious outcomes healthy using diverse cohorts clinical onset assessed using adverse reproductive additional validation 92 %). 2 %), 14 days <p>N-feature model performance metrics. Classification accuracy metrics for models trained on the top three, five, seven, nine, and twelve most important features. Classification metrics of training and testing specimens are shown in Train columns and Test columns. Bootstrapping was used to calculate average values and 95% confidence interval, shown in each cell.</p> 2026-09-15T18:08:25Z Dataset Dataset 10.1371/journal.pcbi.1014768.t002 https://figshare.com/articles/dataset/_p_N-feature_model_performance_metrics_Classification_accuracy_metrics_for_models_trained_on_the_top_three_five_seven_nine_and_twelve_most_important_features_Classification_metrics_of_training_and_testing_specimens_are_shown_in_Train_column/33816697 CC BY 4.0