A mono- and intralink filter (mi-filter) to improve false-discovery rates in cross-linking mass spectrometry data
Cross-Linking Mass Spectrometry (XL-MS) has become an indispensable tool for the emerging field of systems structural biology over the recent years. However, the confidence in individual protein-protein interactions (PPIs) depends on the correct assessment of individual inter protein cross-links. This can be challenging, in particularly in samples where relatively few PPIs are detected, as is often the case in complex samples containing low abundant proteins or in in-vivo settings. In this manuscript we are describing a novel mono- and intralink filter (mi-filter) that is applicable to any kind of crosslinking data and workflow. It stipulates that only proteins for which at least one monolink or intra-protein crosslink has been identified within a given dataset are considered for an inter-protein cross-link and therefore participate in a PPI. We show that this simple and intuitive filter has a dramatic effect on different types of crosslinking-data ranging from single protein complexes, over medium-complexity affinity enrichments to proteome-wide cell lysates and significantly lowers the number of false-positive identifications resulting in improved false-discovery rates for inter-protein links in all these types of XL-MS data.