Docs tagged “validation”
21 Ivy Biosciences docs pages tagged validation.
- How ADMET-AI results compare to the bench
What an ADMET-AI probability has been shown to agree with, why the comparison is one step removed from an assay, and which endpoints do not translate.
- How AiZynthFinder results compare to the bench
What a solved AiZynthFinder route has been shown to agree with experimentally, and why "solved" is a statement about the stock set and template library.
- How AlphaFold results compare to the bench
What an AlphaFold Database model has been shown to agree with experimentally, which AlphaFold version you are holding, and how to read pLDDT and PAE.
- How ANARCI results compare to the bench
What an ANARCI numbering and germline call have been shown to agree with, and the known issue that returns wrong region boundaries under status SUCCESS.
- How BioPhi results compare to the bench
What an OASis score and a Sapiens-humanized sequence have been shown to agree with experimentally, and why humanness does not predict immunogenicity.
- How Boltz-2 results compare to the bench
What a Boltz-2 complex and affinity value have been shown to agree with experimentally, and why a no-MSA single-sample run sits below published figures.
- How Chai-1 results compare to the bench
What a Chai-1 predicted complex has been shown to agree with experimentally, and why a no-MSA single-seed run sits at the weak end of the published range.
- How CRISPR guide results compare to the bench
What a Rule Set 3 on-target score has been shown to agree with, why off-target needs AUC-PR not ROC AUC, and why the near-match scan is not genome-wide.
- How DeepPurpose results compare to the bench
What a predicted drug–target affinity has been shown to agree with experimentally, and why the split a figure was measured under decides what it means.
- How developability results compare to the bench
What a computed developability flag has been shown to agree with experimentally, and why a sequence-motif flag is not a measured degradation rate.
- How DiffDock results compare to the bench
What a DiffDock pose costs against the docking seat and the co-crystal it displaces, and what the confidence score does not tell you.
- How ESM-3 results compare to the bench
What the ESM-3 surfaces on this platform actually run, what the published ESM-3 figures grade, and why the two are not the same model.
- How ESMFold results compare to the bench
What an ESMFold structure has been shown to agree with experimentally, how it compares to MSA-based predictors, and the pose accuracy it costs downstream.
- How GenMol results compare to the bench
What a GenMol-generated molecule has been shown to agree with experimentally, and why every published figure is an in silico benchmark graded by an oracle.
- How Humatch results compare to the bench
What a Humatch humanness score has been shown to agree with experimentally, and what happened when a Humatch-humanized antibody was finally expressed.
- How IgBLAST results compare to the bench
What an IgBLAST germline call has been shown to agree with, why D-gene and allele calls are the weak axis, and which fields this platform cannot compute.
- How MM-GBSA results compare to the bench
What an MM-GBSA binding energy has been shown to agree with experimentally, and why this implementation returns no binding free energy at all.
- How P2Rank results compare to the bench
What a P2Rank pocket has been shown to agree with experimentally, and why almost every published success rate was measured on ligand-bound structures.
- How Paragraph results compare to the bench
What a Paragraph paratope has been shown to agree with experimentally, why a contact is not a binding hot spot, and what the fixed reference chain changes.
- How REINVENT results compare to the bench
What a REINVENT-generated molecule has been shown to agree with experimentally, and why the comparison is makeability rather than accuracy.
- Reading an in silico result against the bench
What a predicted structure, pose or property score has been shown to agree with experimentally, how that agreement is measured, and where the figures stop.