This dataset provides longitudinal phospholipid-class lipidomics in hippocampus and plasma from a transient middle cerebral artery occlusion (tMCAO) model, paired with time-matched sham controls. Lipid class composition is reported as molar percentage (%mol) and includes PA, PG, PI, PS, PE, PC, ether-linked subclasses (ePS, ePE, ePC), lysophospholipids (LPE, LPC), and sphingomyelin (SM). Samples were collected at 6 h, 24 h, 7 d, 15 d, 1 month, and 4 months post-ischemia to capture acute-to-chronic remodeling of membrane lipid composition across tissues and time. Steps to reproduce Download the dataset and open Post-Ischemia Lipidome Trajectory - Nbiol.csv. Subset the data by tissue (hippocampus or plasma) and timepoint (6h, 24h, 7d, 15d, 1m, 4m), including both groups (tMCAO and time-matched sham). Import the %mol lipid species matrix into LipidSig (or an equivalent lipidomics analysis workflow). Zero-filtering: within each tissue × timepoint subset, remove lipid classes/species with >80% zero values across samples. Imputation (timepoint-specific): for each remaining lipid species, replace zero values using the minimum non-zero %mol detected for that species at the same timepoint, computed across sham + tMCAO samples (timepoint-by-timepoint). Perform downstream analyses (e.g., PCA) using the filtered and imputed matrix, as described in the manuscript.