Improved Analysis of Cross-Linking Mass Spectrometry Data with Kojak 2.0, Advanced by Integration into the Trans-Proteomic Pipeline
Michael R. Hoopmann, David Shteynberg, Alex Zelter, Michael Riffle, Andrew S. Lyon, David A. Agard, Qing Luan, Brad J. Nolen, Michael J. MacCoss, Trisha N. Davis, Robert L. Moritz
Abstract
Fragmentation ion spectral analysis of chemically cross-linked proteins is an established technology in the proteomics research repertoire for determining protein interactions, spatial orientation, and structure. Here we present Kojak version 2.0, a major update to the original Kojak algorithm, which was developed to identify cross-linked peptides from fragment ion spectra using a database search approach. A substantially improved algorithm with updated scoring metrics, support for cleavable cross-linkers, and identification of cross-links between 15 N-labeled homomultimers are among the newest features of Kojak 2.0 presented here. Kojak 2.0 is now integrated into the Trans-Proteomic Pipeline, enabling access to dozens of additional tools within that suite. In particular, the PeptideProphet and iProphet tools for validation of cross-links improve the sensitivity and accuracy of correct cross-link identifications at user-defined thresholds. These new features improve the versatility of the algorithm, enabling its use in a wider range of experimental designs and analysis pipelines. Kojak 2.0 remains open-source and multiplatform.