Benchmarking differential expression, imputation and quantification methods for proteomics data

Data analysis is a critical part of quantitative proteomics studies in interpreting biological questions. Numerous computational tools including protein quantification, imputation, and differential expression (DE) analysis were generated in the past decade. However, searching optimized tools is still an unsolved issue. Moreover, due to the rapid development of RNA-Seq technology, a vast number of DE analysis methods are created. Applying these newly developed RNA-Seq-oriented tools to proteomics data is still a question that needs to be addressed. In order to benchmark these analysis methods, a proteomics dataset constituted the proteins derived from human, yeast, and drosophila with different ratios were generated. Based on this dataset, DE analysis tools (including array-based and RNA-Seq based), imputation algorithms, and protein quantification methods were compared and benchmarked. This study provided useful information on analyzing quantitative proteomics datasets. All the methods used in this study were integrated into Perseus which are available at https://www.maxquant.org/perseus.