Skip to main content

Table 3 Classification efficiency of the MixOmics model

From: Influence of cigarette smoking on the human duodenal mucosa-associated microbiota

Training set (leave-one-out validation) FD and ID cohort Test set CD cohort
  Correctly classified (%) Incorrectly classified (%)   Correctly classified (%) Incorrectly classified (%)
Current smoker 93 7 Current smoker 43 57
Never smoker 100 0 Never smoker 83 17
  1. Efficiency of the mixOmics model (sparse partial least squares discriminant analysis) in differentiating patients based on smoking status (current or never) with respect to bacterial profiles. A “training set” which consisted of the FD and ID cohort was used to develop a discriminating model, which was validated using the ‘leave-one-out’ method. The model was then tested on the CD data set