Fig. 1: Study overview. Schematic of the study workflow, from hypotheses (left) to data sources (middle) to analysis steps (right).

a Cross-sectional and longitudinal cortical thickness (CTh) and surface area (SA) data were obtained from three developmental cohorts, including preterm (PT) and full-term (FT) participants. Data were parcellated into 34 bilateral cortical regions. Population CTh and SA life course trajectories were extracted from a normative model. For each participant, individual regional deviations from population life courses were classified as infranormal (i.e., <5th percentile) or supranormal (i.e., >95th percentile) for each cortical region. The regional deviation profile of a certain modality is defined as the measure of the individual brain abnormality pattern (IBAP) for a given participant. IBAP heterogeneity was assessed in two ways: by quantifying the number of subjects with extranormal deviations in each region, and by measuring spatial similarity of regional deviation profiles across subjects. For the latter, binarized regional deviation profiles were cross-correlated to determine the average correlation between each subject’s IBAP with all others. b Consistency in initial brain injury following preterm birth was examined in terms of extent, location, and cellular underpinnings. Extent consistency was examined by relating a subject’s number of extra-normal regional deviations to their gestational age (GA). c Location consistency was exami ned by within-subject longitudinal comparison of IBAPs along childhood and adulthood. d Cellular underpinning consistency was examined by linking gestational age with the spatial correlation of eight brain cell type distributions with adult IBAPs, respectively. Investigated cell types were astrocytes (Astro), endothelial cells (Endo), microglia (Micro), excitatory neurons (Neuro-Ex), inhibitory neurons (Neuro-In), oligodendrocytes (Oligo), oligodendrocyte precursors (OPC), and pericytes (Per). e To investigate the developmental plasticity of IBAPs in children and adults, we captured spatial variability of IBAPs across 34 cortical regions using Principal Component Analysis (PCA) as the main axis of deviation across regions (i.e., PC1) and linked it with social environmental factors during early life.