Abstract / Summary
Cervicovaginal fluid (CVF) represents a valuable source of molecular information reflecting time-dependent proteomic changes across the menstrual cycle. However, the large-scale discovery of CVF biomarkers remains challenging because of limitations in the sensitivity, reproducibility, and data completeness of mass spectrometry-based workflows, particularly within repeated-measures longitudinal study designs. Here, we evaluated a standardised data-independent acquisition (DIA)-based proteomics workflow (dia-PASEF) for longitudinal proteomic analysis of serial CVF samples collected across an ovulation-anchored seven-day window (d0-d6) from a prospective preconception cohort (EARLY-PREG). The performance of this dia-PASEF approach was compared with that of a dda-FracOffline strategy. Across the time window, 2,817 quantifiable proteins were detected with dda-FracOffline, whereas 4,229 quantifiable proteins were detected with dia-PASEF, with mean geometric coefficient of variation values of 203.1% and 55.6%, respectively. Using day 0 (ovulation) as a reference for day-to-day comparisons, more differentially expressed proteins (DEPs) were identified with dia-PASEF than with dda-FracOffline across pairwise comparisons (952-1,116 DEPs for dia-PASEF versus 161-460 DEPs for dda-FracOffline). Pathway analyses demonstrated broader biological coverage with dia-PASEF. In conclusion, dia-PASEF is a sensitive and reproducible approach for longitudinal repeated-measures proteomic analysis of CVF. This method is adaptable to other biofluids and supports the investigation of time-dependent molecular signatures and associated biological pathways under physiologically relevant conditions in female reproductive system research.