rewirebio.iobenchmarks
Contested result

Evo 2 and Caduceus: GUANinE v1.1 reveals complementarity of supervised and genomic language models and GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models disagree

  • Low severity
  • Not yet reviewed
  • two-study-conflict
  • different-task

GUANinE v1.1 reveals complementarity of supervised and genomic language models ranks Caduceus above Evo 2 in 4 of 4 comparisons; GFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models ranks Evo 2 above Caduceus in 35 of 46 comparisons. No other study gives a verdict, so neither is the outlier.

An automatic lead, not a finding of error. The results below are shown as published, everywhere on the site, until a reviewer decides otherwise.

Which studies agree

3 papers rank Evo 2 above Caduceus in 36 of their 58 comparisons and Caduceus higher in 15; 7 are ties. Of those papers, GUANinE v1.1 reveals complementarity… ranks Caduceus higher on its own (4 of 6).

Papers grouped by which system they rank higher: Evo 2 or CaduceusRank Evo 2 higher: GFMBench-API, TraitGym official metrics. Rank Caduceus higher: GUANinE v1.1 reveals complementarity…. Mixed: none.Rank Evo 2 higherMixedRank Caduceus higherGFMBench-APIGFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation ModelsTraitGym official metricsTraitGym official metrics: AUPRC by chromosome, weighted average (Hugging Face dataset songlab/TraitGym)GUANinE v1.1 revealscomplementarity…GUANinE v1.1 reveals complementarity of supervised and genomic language models
  • Rank Evo 2 higher (2): GFMBench-API, TraitGym official metrics
  • Rank Caduceus higher (1): GUANinE v1.1 reveals complementarity…
  • Mixed (0): none

Each paper sits where at least two thirds of its comparisons point; Mixed means neither model reaches two thirds. The larger, outlined dot is the paper that disagrees.

Comparisons in which each paper scores Evo 2 or Caduceus higherGFMBench-API: Evo 2 higher in 35, Caduceus higher in 11, 4 ties. GUANinE v1.1 reveals complementarity…: Evo 2 higher in 0, Caduceus higher in 4, 2 ties. TraitGym official metrics: Evo 2 higher in 1, Caduceus higher in 0, 1 tie.GFMBench-APIGFMBench-API: A Standardized Interface for Benchmarking Genomic Foundation Models35 Evo 2 higher354 tie11 Caduceus higher11GUANinE v1.1 reveals complementarity…GUANinE v1.1 reveals complementarity of supervised and genomic language models2 tie24 Caduceus higher4TraitGym official metricsTraitGym official metrics: AUPRC by chromosome, weighted average (Hugging Face dataset songlab/TraitGym)1 Evo 2 higher11 tie1
Evo 2 higherTieCaduceus higherNumbers are counts of comparisons; the outlined bar is the paper that disagrees.
Comparisons in GFMBench-API (50)
ProtocolDatasetMetricEvo 2CaduceusHigher
GFMBench-API Bend Variant Effects Disease, zs-embeddingBend Variant Effects Disease test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.5960.379Evo 2
GFMBench-API Bend Variant Effects Disease, zs-embeddingBend Variant Effects Disease test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.09440.107Caduceus
GFMBench-API Bend Variant Effects Disease, zs-embeddingBend Variant Effects Disease test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.4740.378Evo 2
GFMBench-API Bend Variant Effects Disease, zs-embeddingBend Variant Effects Disease test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.9120.78Evo 2
GFMBench-API Bend Variant Effects Disease, zs-embeddingBend Variant Effects Disease test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.6150.545Evo 2
GFMBench-API Bend Variant Effects Disease, zs-embeddingBend Variant Effects Disease test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.8720.772Evo 2
GFMBench-API Bend Variant Effects Expression, zs-embeddingBend Variant Effects Expression test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.08970.0692Evo 2
GFMBench-API Bend Variant Effects Expression, zs-embeddingBend Variant Effects Expression test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.08870.0775Evo 2
GFMBench-API Bend Variant Effects Expression, zs-embeddingBend Variant Effects Expression test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.09040.0676Evo 2
GFMBench-API Bend Variant Effects Expression, zs-embeddingBend Variant Effects Expression test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.5730.479Evo 2
GFMBench-API Bend Variant Effects Expression, zs-embeddingBend Variant Effects Expression test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.5570.537Evo 2
GFMBench-API Bend Variant Effects Expression, zs-embeddingBend Variant Effects Expression test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.5820.473Evo 2
GFMBench-API BRCA1, zs-embeddingBRCA1 test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.5750.299Evo 2
GFMBench-API BRCA1, zs-embeddingBRCA1 test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.2040.203Tie
GFMBench-API BRCA1, zs-embeddingBRCA1 test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.540.291Evo 2
GFMBench-API BRCA1, zs-embeddingBRCA1 test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.7750.563Evo 2
GFMBench-API BRCA1, zs-embeddingBRCA1 test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.4670.469Tie
GFMBench-API BRCA1, zs-embeddingBRCA1 test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.7470.555Evo 2
GFMBench-API ClinVar Indel Pathogenicity, zs-embeddingClinVar Indel Pathogenicity test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.9930.959Evo 2
GFMBench-API ClinVar Indel Pathogenicity, zs-embeddingClinVar Indel Pathogenicity test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.9890.958Evo 2
GFMBench-API ClinVar Indel Pathogenicity, zs-embeddingClinVar Indel Pathogenicity test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.9670.71Evo 2
GFMBench-API ClinVar Indel Pathogenicity, zs-embeddingClinVar Indel Pathogenicity test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.9460.708Evo 2
GFMBench-API LOL-EVE, zs-embeddingLOL-EVE test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.09380.0863Evo 2
GFMBench-API LOL-EVE, zs-embeddingLOL-EVE test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.08450.0861Caduceus
GFMBench-API LOL-EVE, zs-embeddingLOL-EVE test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.5430.522Evo 2
GFMBench-API LOL-EVE, zs-embeddingLOL-EVE test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.5310.524Evo 2
GFMBench-API LRB Variant Effect OMIM, zs-embeddingLRB Variant Effect OMIM test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.00210.0019Evo 2
GFMBench-API LRB Variant Effect OMIM, zs-embeddingLRB Variant Effect OMIM test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.0020.0022Caduceus
GFMBench-API LRB Variant Effect OMIM, zs-embeddingLRB Variant Effect OMIM test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.0020.0019Evo 2
GFMBench-API LRB Variant Effect OMIM, zs-embeddingLRB Variant Effect OMIM test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.5020.49Evo 2
GFMBench-API LRB Variant Effect OMIM, zs-embeddingLRB Variant Effect OMIM test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.4980.524Caduceus
GFMBench-API LRB Variant Effect OMIM, zs-embeddingLRB Variant Effect OMIM test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.5010.486Evo 2
GFMBench-API Songlab ClinVar, zs-embeddingSonglab ClinVar test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.8650.991Caduceus
GFMBench-API Songlab ClinVar, zs-embeddingSonglab ClinVar test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.5760.991Caduceus
GFMBench-API Songlab ClinVar, zs-embeddingSonglab ClinVar test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.7340.996Caduceus
GFMBench-API Songlab ClinVar, zs-embeddingSonglab ClinVar test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.8450.867Caduceus
GFMBench-API Songlab ClinVar, zs-embeddingSonglab ClinVar test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.5790.867Caduceus
GFMBench-API Songlab ClinVar, zs-embeddingSonglab ClinVar test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.7220.933Caduceus
GFMBench-API TraitGym Complex, zs-embeddingTraitGym Complex test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.1070.0974Evo 2
GFMBench-API TraitGym Complex, zs-embeddingTraitGym Complex test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.1060.103Evo 2
GFMBench-API TraitGym Complex, zs-embeddingTraitGym Complex test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.1080.0946Evo 2
GFMBench-API TraitGym Complex, zs-embeddingTraitGym Complex test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.5240.497Evo 2
GFMBench-API TraitGym Complex, zs-embeddingTraitGym Complex test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.5230.518Tie
GFMBench-API TraitGym Complex, zs-embeddingTraitGym Complex test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.5270.483Evo 2
GFMBench-API TraitGym Mendelian, zs-embeddingTraitGym Mendelian test set (GFMBench-API)AUPRC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.170.0922Evo 2
GFMBench-API TraitGym Mendelian, zs-embeddingTraitGym Mendelian test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.1210.127Caduceus
GFMBench-API TraitGym Mendelian, zs-embeddingTraitGym Mendelian test set (GFMBench-API)AUPRC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.1640.0879Evo 2
GFMBench-API TraitGym Mendelian, zs-embeddingTraitGym Mendelian test set (GFMBench-API)AUROC (zero-shot; L2 distance of reference and alternate sequence embeddings), unitless0.6260.477Evo 2
GFMBench-API TraitGym Mendelian, zs-embeddingTraitGym Mendelian test set (GFMBench-API)AUROC (zero-shot; cosine similarity of embeddings at the SNV position), unitless0.5480.553Tie
GFMBench-API TraitGym Mendelian, zs-embeddingTraitGym Mendelian test set (GFMBench-API)AUROC (zero-shot; cosine similarity of reference and alternate sequence embeddings), unitless0.6180.463Evo 2
Comparisons in GUANinE v1.1 reveals complementarity… (6)
Comparisons in TraitGym official metrics (2)
ProtocolDatasetMetricEvo 2CaduceusHigher
TraitGym Complex traits, zero-shot, AUPRC by chromosome weighted averageTraitGym Complex traits, matched 9 controlsAUPRC (weighted average over chromosomes of per-chromosome AUPRC; subset 'all'), unitless0.09760.0971Tie
TraitGym Mendelian traits, zero-shot, AUPRC by chromosome weighted averageTraitGym Mendelian traits, matched 9 controlsAUPRC (weighted average over chromosomes of per-chromosome AUPRC; subset 'all'), unitless0.1590.0878Evo 2

What disagrees

studies
2
Compared
Evo 2, Caduceus
Study that disagrees
Other studies
Rule
two-study-conflict; different-task
Model identity
Matched across studies partly through links not yet reviewed

Follow-up

Current state: Not yet reviewed.

No one has reviewed this finding yet.

How to help resolve this

If you know why these studies disagree, for example a different population, metric or preprocessing step, or you have a corrected value, tell us through the contribution form. A reviewer checks every submission against its source before anything changes, and the step is added to the follow-up above.

Finding discrepancy-5b81076ca7ef0584 · All contested results · How findings are made and followed up