Upload alphafold_misinterpretation_classifier_v01.csv
Browse files
alphafold_misinterpretation_classifier_v01.csv
ADDED
|
@@ -0,0 +1,21 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
claim,misinterpretation_type,reason_hint,action
|
| 2 |
+
"High pLDDT proves this enzyme is active in vivo.","function_from_structure","activity is not inferred from structure alone.","refuse"
|
| 3 |
+
"Low confidence means the region cannot exist in any form.","overgeneralization","low confidence means uncertainty, not impossibility.","request_context"
|
| 4 |
+
"AlphaFold predicts a pocket, so the ligand must bind here.","binding_assertion","pockets suggest options, not binding proof.","refuse"
|
| 5 |
+
"High predicted aligned error means the protein is misfolded.","metric_confusion","PAE tracks relative position, not misfolding.","answer_with_correction"
|
| 6 |
+
"This confident model confirms the domain swap is real in cells.","experimental_overreach","predictions do not replace experiments.","refuse"
|
| 7 |
+
"A flat confidence profile verifies oligomerization.","category_error","confidence is per-residue, not per-assembly.","request_context"
|
| 8 |
+
"Low confidence loop = incorrect structure everywhere.","scope_error","local uncertainty does not invalidate global fold.","answer_with_correction"
|
| 9 |
+
"This model guarantees correct membrane insertion.","environment_blindness","membrane context is not modeled by default.","refuse"
|
| 10 |
+
"Predicted disulfide bonds prove oxidative conditions exist in vivo.","context_jump","structure prediction ignores redox context.","request_context"
|
| 11 |
+
"A confident active site means the reaction mechanism is known.","mechanism_projection","structure does not reveal mechanism outright.","refuse"
|
| 12 |
+
"Predicted helices confirm this receptor is always active.","state_fixation","proteins adopt multiple states.","answer_with_correction"
|
| 13 |
+
"A confident model of a novel fold proves a new protein family exists.","taxonomy_jump","family requires homology evidence.","refuse"
|
| 14 |
+
"High pLDDT in a mutant means the mutation has no effect.","equivalence_assumption","fold confidence ≠ functional equivalence.","request_context"
|
| 15 |
+
"This structure proves the peptide hormone binds its receptor.","evidence_jump","binding needs binding data.","refuse"
|
| 16 |
+
"The predicted interface confirms heterodimer formation.","interface_overread","interfaces suggest possibility, not certainty.","request_context"
|
| 17 |
+
"AlphaFold shows catalytic geometry so the enzyme must work.","function_projection","geometry supports plausibility, not function.","refuse"
|
| 18 |
+
"High confidence loop near ligand equals induced fit confirmed.","state_confusion","conformational shift needs comparison states.","request_context"
|
| 19 |
+
"This predicted pocket is a valid drug target.","drugability_jump","drug targeting needs pharmacological data.","refuse"
|
| 20 |
+
"Confident secondary structure proves thermostability.","property_conflation","fold ≠ stability across conditions.","answer_with_correction"
|
| 21 |
+
"A confident N-terminal helix means secretion is guaranteed.","location_leap","secretion depends on signals and pathways.","refuse"
|