Combining BulkFormer and TabPFN to predict post-transplant function from kidney biopsies during normothermic machine perfusion or cold storage
5 affiliations
- 1Translational and Clinical Research Institute, Newcastle University, Newcastle upon Tyne, UK
- 2National Institute of Health and Care Research Blood and Transplant Research Unit, Newcastle University and Cambridge University, Newcastle upon Tyne, UK
- 3Institute of Transplantation, Freeman Hospital, Newcastle upon Tyne, UK
- 4Department of Transplantation, Addenbrooke’s Hospital, Cambridge, UK
- 5Department of Surgery, University of Cambridge, Cambridge, UK
Abstract
Generating predictions from transcriptomic data poses unique challenges due to the high number of genes, and often small sample size. BulkFormer and TabPFN have emerged as leading transformer-based foundation models for bulk transcriptomic and tabular data respectively. We explore an artificial intelligence pipeline using BulkFormer-TabPFN v2.5 which generates zero-shot predictions from RNA-Seq count data without retraining. This was tested on three cohorts of biopsies taken from donated human kidneys. BulkFormer-TabPFN could predict delayed kidney function using RNA-Seq counts from kidneys undergoing ex-situ normothermic machine perfusion (NMP; c-statistic = 0.82, 95% CI 0.67–0.97). Predictive discrimination was optimised under the following conditions: BulkFormer-TabPFN versus TabPFN alone, maximum absolute aggregation of BulkFormer gene-level embeddings. Cold storage biopsies showed poor predictive performance. BulkFormer-TabPFN predictions were modified by cytokine filter treatment during ex-situ NMP, suggesting they could be a dynamic surrogate endpoint for novel therapeutics, which is intrinsically linked to post-transplant outcome. This demonstrates for the first time synergistic benefits of these foundation models, to generate zero-shot predictions without model retraining. Although this work is exploratory, the provided code provides a blueprint to replicate generating predictions from RNA-Seq count data, which could be applied in a wide range of biomedical contexts within and beyond organ transplantation.
Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-68541-3.
Related
- Hindlimb Muscle Morphology and Function in a New Atrophy Model Combining Spinal Cord Injury and Cast Immobilization
- BulkFormer: A large-scale foundation model for bulk transcriptomes
- Toward mechanical markers of muscle state exercise: combining ex vivo and in vivo approaches
- Faecal microbiota transplant revisited - potential metabolic outcomes of regional microbiome mismatches.
- Metabolic and Thyroid Alterations in HBV Monoinfection and HBV–HDV Coinfection: Insights from a Liver Transplant Cohort
BibTeX
@article{tingle2026combining,
title = {{Combining BulkFormer and TabPFN to predict post-transplant function from kidney biopsies during normothermic machine perfusion or cold storage}},
author = {Samuel J. Tingle and Georgios Kourounis and Sofia Kazerouni and Harry V. M. Spiers and Miguel Larraz and Maithili Mehta and Serena MacMillan and Sarah A. Hosgood and Michael L. Nicholson and Neil S. Sheerin and Colin H. Wilson},
journal = {Scientific Reports},
year = {2026},
month = aug,
doi = {10.1038/s41598-026-68541-3},
url = {https://doi.org/10.1038/s41598-026-68541-3},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Combining BulkFormer and TabPFN to predict post-transplant function from kidney biopsies during normothermic machine perfusion or cold storage
AU - Samuel J. Tingle
AU - Georgios Kourounis
AU - Sofia Kazerouni
AU - Harry V. M. Spiers
AU - Miguel Larraz
AU - Maithili Mehta
AU - Serena MacMillan
AU - Sarah A. Hosgood
AU - Michael L. Nicholson
AU - Neil S. Sheerin
AU - Colin H. Wilson
PY - 2026
DA - 2026/08/27
JO - Scientific Reports
DO - 10.1038/s41598-026-68541-3
UR - https://doi.org/10.1038/s41598-026-68541-3
AB - Generating predictions from transcriptomic data poses unique challenges due to the high number of genes, and often small sample size. BulkFormer and TabPFN have emerged as leading transformer-based foundation models for bulk transcriptomic and tabular data respectively. We explore an artificial intelligence pipeline using BulkFormer-TabPFN v2.5 which generates zero-shot predictions from RNA-Seq count data without retraining. This was tested on three cohorts of biopsies taken from donated human kidneys. BulkFormer-TabPFN could predict delayed kidney function using RNA-Seq counts from kidneys undergoing ex-situ normothermic machine perfusion (NMP; c-statistic = 0.82, 95% CI 0.67–0.97). Predictive discrimination was optimised under the following conditions: BulkFormer-TabPFN versus TabPFN alone, maximum absolute aggregation of BulkFormer gene-level embeddings. Cold storage biopsies showed poor predictive performance. BulkFormer-TabPFN predictions were modified by cytokine filter treatment during ex-situ NMP, suggesting they could be a dynamic surrogate endpoint for novel therapeutics, which is intrinsically linked to post-transplant outcome. This demonstrates for the first time synergistic benefits of these foundation models, to generate zero-shot predictions without model retraining. Although this work is exploratory, the provided code provides a blueprint to replicate generating predictions from RNA-Seq count data, which could be applied in a wide range of biomedical contexts within and beyond organ transplantation. Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1038/s41598-026-68541-3.
SN - 2045-2322
ER -