Updated project metadata. This study aimed to investigate the potential of fecal proteomics signatures for NEC prediction. Stools were collected from NEC infants up to 10 days before their diagnosis and matched with those from non-NEC controls for three tested periods. Stool samples were analyzed with liquid chromatography-tandem mass spectrometry using SWATH acquisition and a cross-compatible proteomic software platform to perform label-free quantification. ROC curves and principal component analysis were used to explore discriminating information provided by potential markers and sets of markers were evaluated with classical clustering and classification methods to evaluate their capacity to sort the samples giving to their actual outcomes. We identified a specific series of markers that predicted all NEC from non-NEC infants at least a week in advance. Our study shows that multiplexed proteomic signature detection constitutes a favorable approach for the early detection of NEC.