Assessing rejection-related disease in kidney transplant biopsies based on archetypal analysis of molecular phenotypes
Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
Conventional histologic diagnosis of rejection in kidney transplants has limited repeatability due to its inherent requirement for subjective assessment of lesions, in a rule-based system that does not acknowledge diagnostic uncertainty. Molecular phenotyping affords opportunities for increased precision and improved disease classification to address the limitations of conventional histologic diagnostic systems and quantify levels of uncertainty. Microarray data from 1,208 kidney transplant biopsies were collected prospectively from 13 centers. Cross-validated classifier scores predicting the presence of antibody-mediated rejection (ABMR), T cell-mediated rejection (TCMR), and 5 related histologic lesions were generated using supervised machine learning methods. These scores were used as input for archetypal analysis, an unsupervised method similar to cluster analysis, to examine the distribution of molecular phenotypes related to rejection. Six archetypes were generated: no rejection, TCMR, 3 associated with ABMR (early-stage, fully developed, and late-stage), and mixed rejection (TCMR plus early-stage ABMR). Each biopsy was assigned 6 scores, one for each archetype, representing a probabilistic assessment of that biopsy based on its rejection-related molecular properties. Viewed as clusters, the archetypes were similar to existing histologic Banff categories, but there was 32% disagreement, much of it probably reflecting the "noise" in the current histologic assessment system. Graft survival was lowest for fully developed and late-stage ABMR, and it was better predicted by molecular archetype scores than histologic diagnoses. The results provide a system for precision molecular assessment of biopsies and a new standard for recalibrating conventional diagnostic systems.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.000 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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 it