Research map / Probabilistic and scientific modeling / Molecules and proteins
BayReL: Bayesian Relational Learning for Multi-omics Data Integration
Ehsan Hajiramezanali, Arman Hasanzadeh, Nick Duffield, Krishna Narayanan, Xiaoning Qian
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use BayReL to infer cross-view interactions instead of BCCA | CF microbiome-metabolome interaction inference · CF | positive accuracy at negative accuracy >97%: 28.30% ± 3.21 -> 82.7% ± 4.7 | Improved |
| Use BayReL to infer cross-view interactions instead of SRCA | CF microbiome-metabolome interaction inference · CF | positive accuracy at negative accuracy >97%: 26.41% -> 82.7% ± 4.7 | Improved |
| Use BayReL instead of BCCA | TCGA BRCA interaction inference · TCGA BRCA, avg. degree 0.2, 0.3, and 0.4 | prediction sensitivity: 21.08 ± 0.0 -> 34.06 ± 2.5; 31.18 ±0.7 -> 47.46 ± 2.6; 41.12 ±0.2 -> 59.50 ± 3.0 | Improved |
| Use BayReL instead of SRCA | TCGA BRCA interaction inference · TCGA BRCA, avg. degree 0.2, 0.3, and 0.4 | prediction sensitivity: 17.58 -> 34.06 ± 2.5; 28.26 -> 47.46 ± 2.6; 37.55 -> 59.50 ± 3.0 | Improved |
| Use BayReL instead of BCCA | AML gene-drug interaction inference · AML, avg. degrees 0.10–0.50 | prediction sensitivity: 15.56 ± 0.75 vs 9.65 ±0.75; 21.70 ± 0.65 vs 14.34 ±0.06; 27.20 ± 0.17 vs 18.96 ±0.42; 32.43 ± 1.02 vs 23.29 ±0.52; 37.76 | Improved |
| Use BayReL instead of SRCA | AML gene-drug interaction inference · AML, avg. degrees 0.10–0.50 | prediction sensitivity: 8.03 -> 15.56 ± 0.75; 12.00 -> 21.70 ± 0.65; 17.15 -> 27.20 ± 0.17; 20.70 -> 32.43 ± 1.02; 26.85 -> 37.76 ± 0.85; 34.93 | Improved |
| Train with 50% rather than 100% of samples | BayReL on TCGA BRCA · TCGA BRCA, avg. degrees 0.20, 0.30, and 0.40 | prediction sensitivity: 34.0 -> 32.1; 47.4 -> 45.9; 59.5 -> 58.7 | Hurt |
| Train with 50% rather than 100% of samples | BCCA on TCGA BRCA · TCGA BRCA, avg. degrees 0.20, 0.30, and 0.40 | prediction sensitivity: 21.0 -> 17.6; 31.1 -> 26.4; 41.1 -> 35.5 | Hurt |
| Use 25% rather than 100% of samples | BayReL on TCGA BRCA · TCGA BRCA, avg. degrees 0.20, 0.30, and 0.40 | prediction sensitivity: 34.0 -> 31.9; 47.4 -> 45.8; 59.5 -> 57.6 | Hurt |
| Use 25% rather than 100% of samples | BCCA on TCGA BRCA · TCGA BRCA, avg. degrees 0.20, 0.30, and 0.40 | prediction sensitivity: 21.0 -> 17.4; 31.1 -> 26.0; 41.1 -> 35.4 | Hurt |
| Infer interactions from AML patient samples rather than cell-line samples | BayReL · AML patient and cell-line inferred bipartite graphs | KL divergence: 0.66 -> 0.38 | Improved |
| Infer interactions from fewer training samples | BayReL on TCGA BRCA · TCGA BRCA inferred bipartite graph consistency | KL divergence: BayReL: 0.35 and 0.32; BCCA: 0.67 and 0.62 | Improved |
Later papers that compared with this method
1 later accepted paper reports BayReL as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- MoReL: Multi-omics Relational Learning (ICLR 2022 · better on 4 of 4 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Knowledge Enhanced Representation Learning for Drug Discovery (AAAI 2024 · similarity 0.84)
- PRAGA: Prototype-aware Graph Adaptive Aggregation for Spatial Multi-modal Omics Analysis (AAAI 2025 · similarity 0.83)
- Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task Datasets (ICLR 2024 · similarity 0.83)
- Approximating mutual information of high-dimensional variables using learned representations (NeurIPS 2024 · similarity 0.82)
- FlexMol: A Flexible Toolkit for Benchmarking Molecular Relational Learning (NeurIPS 2024 · similarity 0.82)
- Learning Molecular Representation in a Cell (ICLR 2025 · similarity 0.82)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
