BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
Noncoding variant effects on expression. Scored with AUROC on DeepSEA expression variants. A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1.
Overview
Noncoding variant effects on expression. Scored with AUROC on DeepSEA expression variants. A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1.
Consult the linked sources for architecture or protocol details. Missing evidence is not evidence of a missing capability.
Evaluation design
Benchmarks bring together tasks and protocols. A task describes the biological question; a protocol defines a particular test.
Benchmarks
These source-backed links do not make different protocols or scores interchangeable.
Recorded evaluations
Each evaluation records what was tested and under which conditions.
- AWD-LSTM on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- DEEPSEA on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- DNABERT-2 on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- DNABERT on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- GENA-LM BERT on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- GENA-LM BigBird on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- GROVER on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- HyenaDNA large on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- HyenaDNA tiny on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- NT-1000G on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- NT-H on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- NT-MS on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
Run instructions
No runnable recipe has been reviewed for this task. Dataset access, model requirements, licences and compute requirements must be checked against its sources before execution.
A task describes a biological question. Choose a linked protocol to obtain concrete split and scoring instructions.
Published comparisons
Explore the results reported under one evaluation protocol. Each figure keeps its source, dataset and metric together; it is not a ranking across studies. The pooled view gathers every source table that reports the same metric and names what it does not hold constant.
BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
auroc (fraction) · Higher values are better for this metric.
Every method BEND reports on Noncoding variant effects on expression, scored with AUROC on DeepSEA expression variants.
Evaluation protocol · DeepSEA expression variants (BEND split)
- DEEPSEA · Configuration · Author-reported evaluation0.70
- ResNet-LM · Method · Author-reported evaluation0.55
- AWD-LSTM · Method · Author-reported evaluation0.53
- NT-H · Method · Author-reported evaluation0.55
- NT-MS · Method · Author-reported evaluation0.54
- NT-1000G · Configuration · Author-reported evaluation0.45
- NT-V2 · Configuration · Author-reported evaluation0.48
- DNABERT · Configuration · Author-reported evaluation0.60
- DNABERT-2 · Configuration · Author-reported evaluation0.49
- GENA-LM BERT · Configuration · Author-reported evaluation0.49
- GENA-LM BigBird · Configuration · Author-reported evaluation0.49
- HyenaDNA large · Configuration · Author-reported evaluation0.51
- HyenaDNA tiny · Configuration · Author-reported evaluation0.47
- GROVER · Configuration · Author-reported evaluation0.56
Source order is preserved. Plotted marks show point estimates; uncertainty, where reported, is retained in the printed values and table. Differences do not establish statistical significance.
BEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 1, row(Noncoding variant effects on expression)Values, uncertainty and evidence
| Tested entity | Printed value | Uncertainty | Evidence |
|---|---|---|---|
| DEEPSEA · Configuration | 0.70 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(Expert method), column(Noncoding variant effects on expression) |
| ResNet-LM · Method | 0.55 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(ResNet-LM), column(Noncoding variant effects on expression) |
| AWD-LSTM · Method | 0.53 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(AWD-LSTM), column(Noncoding variant effects on expression) |
| NT-H · Method | 0.55 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-H), column(Noncoding variant effects on expression) |
| NT-MS · Method | 0.54 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-MS), column(Noncoding variant effects on expression) |
| NT-1000G · Configuration | 0.45 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-1000G), column(Noncoding variant effects on expression) |
| NT-V2 · Configuration | 0.48 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-V2), column(Noncoding variant effects on expression) |
| DNABERT · Configuration | 0.60 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(DNABERT), column(Noncoding variant effects on expression) |
| DNABERT-2 · Configuration | 0.49 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(DNABERT-2), column(Noncoding variant effects on expression) |
| GENA-LM BERT · Configuration | 0.49 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(GENA-LM BERT), column(Noncoding variant effects on expression) |
| GENA-LM BigBird · Configuration | 0.49 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(GENA-LM BigBird), column(Noncoding variant effects on expression) |
| HyenaDNA large · Configuration | 0.51 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(HyenaDNA large), column(Noncoding variant effects on expression) |
| HyenaDNA tiny · Configuration | 0.47 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(HyenaDNA tiny), column(Noncoding variant effects on expression) |
| GROVER · Configuration | 0.56 fraction | Not reported | Author-reported evaluation · source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(GROVER), column(Noncoding variant effects on expression) |
Scope and limitations
- Author-reported numbers, source checked but not independently reproduced.
- The expert entries are specialist published models, each compared on one task only.
- The metric differs by task, taken from Table 1, so these figures cannot be averaged into one score.
Source transcription and grouping reviewed by automated source review on 2026-09-18. These experiments were not independently reproduced by rewire.
Tested entities and results
Release 2026-09-17-134cd1815de8 · 14 evaluations · 14 metric rows. Different protocols are not a single leaderboard. Where several source tables report the same metric, the published comparisons above offer a pooled view that names what it does not hold constant.
| Metric and finding | Coverage and uncertainty | Evidence |
|---|---|---|
| AWD-LSTM on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Method: AWD-LSTMTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.53 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(AWD-LSTM), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| DEEPSEA on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: DEEPSEATask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.70 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(Expert method), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| DNABERT-2 on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: DNABERT-2Task: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.49 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(DNABERT-2), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| DNABERT on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: DNABERTTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.60 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(DNABERT), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| GENA-LM BERT on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: GENA-LM BERTTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.49 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(GENA-LM BERT), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| GENA-LM BigBird on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: GENA-LM BigBirdTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.49 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(GENA-LM BigBird), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| GROVER on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: GROVERTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.56 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(GROVER), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| HyenaDNA large on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: HyenaDNA largeTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.51 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(HyenaDNA large), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| HyenaDNA tiny on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: HyenaDNA tinyTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.47 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(HyenaDNA tiny), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| NT-1000G on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: NT-1000GTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.45 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-1000G), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| NT-H on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Method: NT-HTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.55 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-H), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| NT-MS on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Method: NT-MSTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.54 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-MS), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| NT-V2 on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Configuration: NT-V2Task: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.48 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(NT-V2), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
| ResNet-LM on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression Method: ResNet-LMTask: BEND VARIANT-EXPRESSION: Noncoding variant effects on expressionDataset subset: DeepSEA expression variants (BEND split) A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1. Author-reported evaluation · Evaluation metadata: source checked | ||
| 0.55 auroc Unit: fraction · Direction: higher | Uncertainty: Not reported Scored: Not reported · Eligible: Not reported | source checkedBEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Table 3, row(ResNet-LM), column(Noncoding variant effects on expression) Source checking is not independent reproduction. |
Evidence table
Inspect claims, sources and review details
Trace each statement to its source and review. A context-only reference supports the record generally; it does not verify an individual field. Source checking does not reproduce an experiment.
One row per statement and cited source. Multiple citations are not independent evaluations. Shared locators are labelled explicitly.
1 evidence row matching the loaded filters
| Property and statement | Original source and location | Review and provenance |
|---|---|---|
| Relationship: part of discovery-benchmark-bend Individual claims | BEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks Table 1, row(Noncoding variant effects on expression) Version: Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256 | source checked automated source review · 2026-09-18 Audit detailsPrimary-source transcription with no human sign-off and no independent reproduction. Field: Claim: bend-association-variant-expression Source artifact SHA-256: Hash scope: Exact retrieved primary paper artifact bytes. |
Sources and history
Release 2026-09-17-134cd1815de8 · Record review: source checked
1 source records and release history
- BEND: Benchmarking DNA Language Models on Biologically Meaningful Tasks · Original source · Primary full-text snapshot retrieved 2026-09-17; exact bytes pinned by SHA-256
Technical metadata and extraction receipts
Stable ID: bend-task-variant-expression
- areas
- dna-genomes
- tasks
- Noncoding variant effects on expression
- metric
- AUROC
- metric direction
- higher
- dataset
- DeepSEA expression variants
- protocol
- A downstream head trained on frozen embeddings, except for the expert methods and the fully supervised baselines, which are trained end to end. Metric and splits are from Table 1.
- source locator
- Table 1, row(Noncoding variant effects on expression)
- comparison panels
- id: bend-panel-variant-expression; title: BEND VARIANT-EXPRESSION: Noncoding variant effects on expression; protocol id: bend-task-variant-expression; dataset id: bend-dataset-deepsea-expression-variants; metric: auroc; unit: fraction; direction: higher; result ids: bend-result-deepsea-variant-expression-auroc; bend-result-resnet-lm-variant-expression-auroc; bend-result-awd-lstm-variant-expression-auroc; bend-result-nt-h-variant-expression-auroc; bend-result-nt-ms-variant-expression-auroc; bend-result-nt-1000g-variant-expression-auroc; bend-result-nt-v2-variant-expression-auroc; bend-result-dnabert-variant-expression-auroc; bend-result-dnabert-2-variant-expression-auroc; bend-result-gena-lm-bert-variant-expression-auroc; bend-result-gena-lm-bigbird-variant-expression-auroc; bend-result-hyenadna-large-variant-expression-auroc; bend-result-hyenadna-tiny-variant-expression-auroc; bend-result-grover-variant-expression-auroc; source ids: evidence-expansion-bend-final-f709b6be; source locator: Table 1, row(Noncoding variant effects on expression); context: Every method BEND reports on Noncoding variant effects on expression, scored with AUROC on DeepSEA expression variants.; caveats: Author-reported numbers, source checked but not independently reproduced.; The expert entries are specialist published models, each compared on one task only.; The metric differs by task, taken from Table 1, so these figures cannot be averaged into one score.; review: method: automated_source_review; date: 2026-09-18
Related records
- part of: BEND
- subject: BEND VARIANT-EXPRESSION: part of discovery-benchmark-bend
- benchmark: AWD-LSTM on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: DEEPSEA on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: DNABERT-2 on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: DNABERT on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: GENA-LM BERT on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: GENA-LM BigBird on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: GROVER on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: HyenaDNA large on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: HyenaDNA tiny on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: NT-1000G on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: NT-H on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: NT-MS on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: NT-V2 on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression
- benchmark: ResNet-LM on BEND VARIANT-EXPRESSION: Noncoding variant effects on expression