SERRF - sytematical error removal using random forest.

a QC-based normalization method for large-scale untargeted metabolomics.

online

• address: http://serrf.fiehnlab.ucdavis.edu
• functionality:

1. use quality control (QC) samples to build SERRF to normalize biology samples and validate samples.
2. calculate relative standard deviation (RSD) on the validate samples to access performance quantitatively.
3. perform principal component analysis (PCA) to access performance visually.

more tutorial and explanation available online. The online version have only be tested in a limited number of data sets. If any error/suggestion, please feel free to contact me at [email protected]. I'll try my best to answer your question within 24 hours.

citation (being submitted)

Sili Fan, Tomas Cajka, Stanley L. Hazen, W.H. Wilson T ang, Dinesh K. Barupal, and Oliver Fiehn. Systematical Error Removal using Random Forest (SERRF) for large-scale untargeted metabolomics