Cluster analysis of transcriptomic datasets to identify endotypes of idiopathic pulmonary fibrosis
Bibliographic record
Abstract
Background Considerable clinical heterogeneity in idiopathic pulmonary fibrosis (IPF) suggests the existence of multiple disease endotypes. Identifying these endotypes would improve our understanding of the pathogenesis of IPF and could allow for a biomarker-driven personalised medicine approach. We aimed to identify clinically distinct groups of patients with IPF that could represent distinct disease endotypes. Methods We co-normalised, pooled and clustered three publicly available blood transcriptomic datasets (total 220 IPF cases). We compared clinical traits across clusters and used gene enrichment analysis to identify biological pathways and processes that were over-represented among the genes that were differentially expressed across clusters. A gene-based classifier was developed and validated using three additional independent datasets (total 194 IPF cases). Findings We identified three clusters of patients with IPF with statistically significant differences in lung function (p=0.009) and mortality (p=0.009) between groups. Gene enrichment analysis implicated mitochondrial homeostasis, apoptosis, cell cycle and innate and adaptive immunity in the pathogenesis underlying these groups. We developed and validated a 13-gene cluster classifier that predicted mortality in IPF (high-risk clusters vs low-risk cluster: HR 4.25, 95% CI 2.14 to 8.46, p=3.7×10 −5 ). Interpretation We have identified blood gene expression signatures capable of discerning groups of patients with IPF with significant differences in survival. These clusters could be representative of distinct pathophysiological states, which would support the theory of multiple endotypes of IPF. Although more work must be done to confirm the existence of these endotypes, our classifier could be a useful tool in patient stratification and outcome prediction in IPF.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.003 | 0.007 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.002 |
| Bibliometrics | 0.006 | 0.005 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.000 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".