This data set includes a total of 977 de-identified subjects and 75 metabolites without missing values. These metabolites include free fatty acids, amino acids, and bile acids, which were identified using both GC/MS-based non-targeted analysis and LC/MS-based targeted metabolomics approach. It served as a large sample size data set for label-free evaluation.
This data set was collected from a study of comparing metabolic profiles between obese subjects with diabetes mellitus and healthy controls28 (link),29 (link). After filtering all missing values, this data set contained a total number of 198 subjects (70 patients, 128 healthy controls) and 130 metabolites. These metabolites include free fatty acids, amino acids, and bile acids that were identified using LC/MS-based targeted metabolomics approaches. It served as medium sample size data set for both label-free and labeled data evaluation.
Then the other two datasets with missing elements were applied to determine the types of missing values present in different metabolomics datasets.
The is a GC/MS profiling data that contains 37 samples and 110 metabolites identified, with 317 missing values and 221 of them were re-identified manually.
This is a targeted LC/MS metabolomics dataset, which includes 40 samples and 41 metabolites, with 88 missing elements and 26 of them were re-identified manually.