Notice: The reproducibility variables underlying each score are classified using an automated LLM-based pipeline, validated against a manually labeled dataset. LLM-based classification introduces uncertainty and potential bias; scores should be interpreted as estimates. Full accuracy metrics and methodology are described in Coakley et alK. L. Coakley, T. Snelleman, H. Hoos, and O. E. Gundersen, "The embrace of open science: An analysis of a decade of AI research and 56 800 conference papers," Under Review, 2026..
Spatially Resolved Gene Expression Prediction from Histology Images via Bi-modal Contrastive Learning
Authors: Ronald Xie, Kuan Pang, Sai Chung, Catia Perciani, Sonya MacParland, Bo Wang, Gary Bader
NeurIPS 2023 | Venue PDF | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | We demonstrate BLEEP s effectiveness in gene expression prediction by benchmarking its performance on a human liver tissue dataset captured using the 10x Visium platform, where it achieves significant improvements over existing methods. |
| Researcher Affiliation | Academia | 1University of Toronto, 2Vector Institute, 3University Health Network, 4The Donnelly Centre, 5Toronto General Hospital Research Institute, 6Canadian Institute for Advanced Research (CIFAR) |
| Pseudocode | Yes | Algorithm 1 Bimodal Embedding for Expression Prediction |
| Open Source Code | Yes | Code available at https://github.com/bowang-lab/BLEEP |
| Open Datasets | Yes | The dataset [2, 3] used to train and benchmark BLEEP... The data is publicly available for download at https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE240429. |
| Dataset Splits | No | The paper mentions holding out one slice for testing and using the others for training, but it does not explicitly state specific percentages or methods for a separate validation split within the main text. |
| Hardware Specification | Yes | BLEEP is trained using 4 NVIDIA V100 GPUs with the Adam W optimizer[12], a batch size of 512 and a learning rate of 0.001 for 150 epochs. |
| Software Dependencies | No | The paper mentions software like "Scanpy package" and "Harmony", but it does not provide specific version numbers for these software dependencies, which is required for reproducibility. |
| Experiment Setup | Yes | BLEEP is trained using 4 NVIDIA V100 GPUs with the Adam W optimizer[12], a batch size of 512 and a learning rate of 0.001 for 150 epochs. |