RPML: A Learning-Based Approach for Reranking Protein-Spectrum Matches

摘要

Searching top-down spectra against a protein database has been a mainstream method for intact protein identification. Ranking true Protein-Spectrum Matches (PrSMs) over their false counterparts is a feasible method for improving protein identification results. In this paper, we propose a novel model called RPML (Rerank PrSMs based on Machine Learning) to rerank PrSMs in top-down proteomics. The experimental results on real data sets show that RPML can distinguish more correct PrSMs from incorrect ones. The source codes of algorithm are available at.