AI Science Tracker
EN

DeepMILO: a deep learning approach to predict the impact of non-coding sequence variants on 3D chromatin structure

Affiliation not stated in the reviewed material · 26 Mar 2020 · medicine

Discovery

DeepMILO: a deep learning approach to predict the impact of non-coding sequence variants on 3D chromatin structure

Source check: Source checked. Not independent scientific confirmation.

Finding: This method paper reports a specific cancer-genomics result (recurrent predicted loop disruption and association with BCL2/MYC expression), supported by learned sequence modeling and genomic expression analysis. Keep associations and predictions provisional. Methods/Results: CNN/RNN learn DNA sequence features for CTCF anchor/orientation and loop predictions. Test AUPRC varies by negative set (0.684 hardest); application to WGS of 1,834 patients across 12 cancer types predicts 672 insulator loops disrupted in ≥10% of patients; variants at anchors associated with BCL2/MYC upregulation in malignant lymphoma. Association/predictions computational, with validation on published loop datasets/deletion examples; does not prove causal mechanism.

Technology: CNN/RNN learn DNA sequence features for CTCF anchor/orientation and loop predictions. Test AUPRC varies by negative set (0.684 hardest); application to WGS of 1,834 patients across 12 cancer types predicts 672 insulator loops disrupted in ≥10% of patients; variants at anchors associated with BCL2/MYC upregulation in malignant lymphoma. Association/predictions computational, with validation on published loop datasets/deletion examples; does not prove causal mechanism.

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
CNN/RNN learn DNA sequence features for CTCF anchor/orientation and loop predictions. Test AUPRC varies by negative set (0.684 hardest); application to WGS of 1,834 patients across 12 cancer types predicts 672 insulator loops disrupted in ≥10% of patients; variants at anchors associated with BCL2/MYC upregulation in malignant lymphoma. Association/predictions computational, with validation on published loop datasets/deletion examples; does not prove causal mechanism.
Test and result
CNN/RNN learn DNA sequence features for CTCF anchor/orientation and loop predictions. Test AUPRC varies by negative set (0.684 hardest); application to WGS of 1,834 patients across 12 cancer types predicts 672 insulator loops disrupted in ≥10% of patients; variants at anchors associated with BCL2/MYC upregulation in malignant lymphoma. Association/predictions computational, with validation on published loop datasets/deletion examples; does not prove causal mechanism.
Limitation / still unknown
Methods/Results: CNN/RNN learn DNA sequence features for CTCF anchor/orientation and loop predictions. Test AUPRC varies by negative set (0.684 hardest); application to WGS of 1,834 patients across 12 cancer types predicts 672 insulator loops disrupted in ≥10% of patients; variants at anchors associated with BCL2/MYC upregulation in malignant lymphoma. Association/predictions computational, with validation on published loop datasets/deletion examples; does not prove causal mechanism. The source does not establish benefits beyond the reported evaluation.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://link.springer.com/article/10.1186/s13059-020-01987-4

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: other. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 2/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.

Related catalog entries

Links use documented relationships where available, plus shared catalog context. A shared topic or model family does not establish a scientific dependency.