Match Between Runs

A foundational algorithm for many quantitative pipelines is ‘match-between-runs’, which improves the sensitivity and completeness of protein quantitation. The match-between-runs (MBR) algorithm was designed for bulk proteomics data and is being liberally applied to single cell proteomics (SCP) data. Professors Sam Payne and Ryan Kelly at BYU recently discovered that a basic assumption of MBR is not met for single cell proteomics. Therefore its application to SCP datasets leads to significant inaccuracies.

Fortunately, we have discovered that SCP data contains additional information that can be incorporated into MBR to improve the accuracy of protein quantitation. Thus we have improved the MBR algorithm in MetaMorpheus to transform it for SCP data. This algorithm update is achieved through addition of a spectrum validation component that is absent from other, current MBR algorithms. For each MBR peak, we return to the raw data and search for tandem mass spectra that potentially containing fragment peaks from the corresponding peptide. We compare library spectra to these experimental spectra and measure spectral similarity. Armed with this new information, we compute a statistical confidence metric for the MBR peak that users can consider when including MBR data in their quantitative experiments. Importantly, many MBR peaks are false positives and many available MBR algorithms fail to consider this. The spectral library component of MetaMorpheus' MBR algorithm goes a long way toward eliminating many false positive peaks.

Protocol for performing MBR with spectrum similarity validation in MetaMorpheus

Description of information contained in the MBR output