Research On Pre-processing Methods For Label-free Quantitative Proteomics Data With Large Sample

Based on the feature of label-free quantitative proteomics data with large sample and the “lag” of data pre-processing methods in the field of proteomics,R package preprocessor which is specialized in the preprocess of label-free proteomics data with large sample was developed for the rapidly growing demands of experimental data pre-processing with large sample brought by the extraordinary growth of precision medicine in this research.After filtering the data pre-processing methods in other omics research fields,preprocessor has integrated some pre-processing methods appropriate for label-free proteomics data with large sample,such as EigenMS,QCRLSC and missForest,etc.These methods can handle the problem of data bias caused by large sample and high heterogeneity in clinical sample as well as the problem of data missing brought by label-free quantitative methods.At the same time,preprocessor have embodied classical data pre-processing methods of proteomics for convenient comparison and conservative selection.This package has comprehensive work-flow of data pre-processing for label-free quantitative proteomics with large sample,containing data reading,cleaning,pre-assessment,normalization,imputation,the comparison of these methods and so on,moreover,it can generate reports of these pre-processing results.So preprocessor could assist researchers to find more suitable pre-processing methods and get over specific difficulties faced by label-free quantitative proteomics with large sample.