Skip to main content
Fig. 2 | BMC Bioinformatics

Fig. 2

From: Random forest-based imputation outperforms other methods for imputing LC-MS metabolomics data: a comparative study

Fig. 2

Imputation work flow. In every dataset from the dataset grid (12 datasets), which is randomly selected, the missing values are filtered out and 200 molecular features are randomly chosen. Then seven different missing mechanisms are simulated; Missing Completely At Random (MCAR), Missing At Random (MAR), Missing Not At Random (MNAR), MCAR-MAR, MCAR-MNAR, MNAR-MAR, MNAR-MCAR-MAR, in four different percentages (5, 10, 20, and 30%) of missing data. In every dataset that is chosen randomly, nine imputation are used in order to investigate the performance of the imputation methods in estimating missing values. The evaluation of the methods is done using NRMSE. The whole processes are repeated 100 times

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us