A Straightforward Interpretation of Proximity Labeling through Direct Biotinylation Analysis

Proximity labeling (PL) is a revolutionary tool in proteomics,
enabling the precise identification of protein interactions in live
cells. However, conventional statistical approaches for analyzing
biotinylation data often lead to false positives, hindering the accuracy
of the PL studies. In this study, we propose a direct biotinylation
analysis approach that focuses on identifying only biotinylated peptides
rather than relying solely on statistical comparisons. Using LC-MS
data from a prior TurboID-based study, we reanalyzed secretome data
sets and demonstrated significant improvements in identifying true
biotinylated proteins with fewer false positives. By applying this
approach to tissue-specific secretome data, we identified fibronectin
(FN1) as a pericyte-specific marker. Our findings highlight that the
limitations of traditional methods are insufficiently robust, and
we advocate for the adoption of direct biotinylation analysis to enhance
data reliability in PL-based proteomics. This methodology sets a new
standard for studying protein interactions and secretomes, offering
deeper insights into cellular- and tissue-specific molecular networks.