Conserved microglial programmes, non-transferable classifiers: boundaries of the single-cell TBI molecular clock
Listed in
This article is not in any list yet, why not save it to one of your lists.Abstract
Background
Traumatic brain injury (TBI) pathology evolves over hours to months, and molecular staging could complement clinical and imaging-based assessment. Single-cell transcriptomics captures cell-type–resolved temporal programs, but whether a transcriptional “molecular clock” trained on one injury model can be transferred to others remains unknown.
Methods
We compiled four public mouse TBI single-cell/single-nucleus datasets spanning two injury models — controlled cortical impact (GSE277487: 162,892 cells; CEREBRI [GSE269748]: 39,026 nuclei) and fluid percussion injury (GSE160763: 20,642 cells) — plus an independent mild-FPI cohort (mTBI) reserved for validation (GSE247339: 85,431 cells; vehicle-treated arm of a pharmacological study), totalling 307,991 cells across four pathological stages (control, acute, subacute, chronic). Random-forest classifiers on Harmony-corrected principal components were evaluated at both cell and sample levels (held-out 10x libraries), with significance tested against a permutation null that shuffles the sample–stage mapping; we further assessed per-cell-type performance, strict leave-one-dataset-out (LODO) generalization with fully train-only preprocessing, glial-only cross-dataset transfer, and gene-level SHAP attribution.
Results
Within GSE277487, sample-level balanced accuracy reached 0.67 (permutation null 0.32 ± 0.09; p = 0.001), and the signal was reproduced in the independent mild-FPI cohort (0.40 vs null 0.19 ± 0.05; p = 0.0010). Cell-level cross-validation overestimated performance by ∼0.16, quantifying within-sample leakage. Cross-dataset generalization was poor (strict LODO balanced accuracy 0.20-0.31); the acute immediate-early response was untransferable (recall 0.002) and 80% of chronic (6-month) cells were assigned to the subacute class. Glial-only models raised subacute recall across models (0.91 and 0.75), but precision was indistinguishable from a subacute-only classifier (0.51 vs 0.50; 0.58 vs 0.58) and one-vs-rest AUROC was at or below chance (0.38-0.54) across four classifier families — classifier collapse, not transferable discrimination. The reproducible cross-model component was instead a gene-expression programme: a microglial complement/lysosomal damage programme (Trem2-Apoe axis) was up-regulated across the severe CCI and FPI models (13/13 concordant genes), whereas the immediate-early module was model-specific.
Conclusions
A single-cell molecular clock for TBI staging is feasible within datasets once within-sample leakage is controlled, but it does not transfer across datasets as a classifier: apparent cross-model transfer reflected classifier collapse onto the majority class, and the reproducible cross-model component was a microglial complement/lysosomal gene-expression programme. These findings define the boundaries of cross-dataset time inference in TBI and provide evaluation criteria for claims of transferable single-cell classifiers.
