Python implementation of the Systematic Error Removal Using Random Forest (SERRF) algorithm. SERRF is a qc-based sample normalization method designed for large-scale untargeted metabolomics data. The method was developed by the Fan et al. in 2015 (see https://slfan2013.github.io/SERRF-online/). This is simply an attempt to port its functionality from R to python. The package structure is based on SKlearn's transformers, with fit and transform methods.
Documentation can be found at https://pyserrf.readthedocs.io
TODO: Implement cross-validation (almost done) TODO: Verify if injection time is accounted for with current code TODO: Add documentation TODO: Add more tests TODO: Add CLI