BoltzMaker

ABL1_KD packed 2026-08-23

1 Proteins
0 Co-folded partners
2 Ligands
1 Pockets
1 Apo structure references
5 Predictions

Campaign summary

Field Value Details
Input file boltz_input.md boltz_input.md
Proteins 1 2 protein block(s) in 1 group(s): ABL1 (272 aa); ABLAP (272 aa)
Co-folded partners 0 none
Ligands 2 IMATI (SMILES); DASAT (SMILES)
Pockets 1 1N1
Apo structure references 1 ABLAP
Predictions 5 ABL1_IMATI, ABL1_IMATI_1N1, ABL1_DASAT, ABL1_DASAT_1N1, ABLAP
Predict affinity yes pIC50 predicted for every ligand-bound prediction
Ligand chemistry 2 of 2 flagged IMATI, DASAT -- see "Ligand preparation" below
Boltz predict runtime 18m 11s across 2 run invocations
Accelerator gpu gpu = Metal/CUDA backend used; cpu = no GPU available
Workers 0 parallel data-loading workers (Boltz's own default is 2)
MPS watermark 1.0 PYTORCH_MPS_HIGH_WATERMARK_RATIO cap -- lower avoids swap on Apple unified memory
Max parallel samples 1 Boltz's own --max_parallel_samples
Max MSA sequences 4096 cap on MSA sequences used for co-evolution features

Reference structures

The experimental structures this campaign was built on: where each pocket came from, and what the secondary-structure comparison measures against.

Pocket definitions

PocketFromProteinsContacts
1N12GQGABL118
STI1IEPABL1reference only

Secondary-structure references

ProteinStructureStateContentsChain
ABL12gqg.cifDFG-in, αC-in2 chain(s), bound: 1N1, PTRA

A kinase's state is its DFG motif and αC helix, taken from the secondary-structure comparison.

Ligand definitions

LigandClassGiven asExperimental structure
IMATIControlSMILESSTI
DASATControlSMILES1N1

A control has been verified experimentally, by structure or by assay, so its prediction can be checked; an experimental compound is under investigation, with nothing to check against. Experimental compounds are ringed in red in the charts.

pIC50 vs confidence score

Click a point to open that target, in the panel below the Targets table.

pIC50 vs binder probability

Click a point to open that target, in the panel below the Targets table.

Ranked predicted pIC50

Click a point to open that target, in the panel below the Targets table.

Ranked confidence

Click a point to open that target, in the panel below the Targets table.

Summary table

IdentityAffinity Confidence (5) Interactions (5)Structure
RunProteinLigandPocketClassSummaryBinderpIC50ScorepTMipTMLigpLDDTTotalPhobHSaltπCIF
4ABL1DASAT1N1Control 0.9912.860.950.970.980.980.9410.007300CIF
2ABL1IMATI1N1Control 0.8110.100.950.960.980.980.9416.0010411CIF
3ABL1DASATUncControl 0.559.890.880.690.880.880.8811.008300CIF
1ABL1IMATIUncControl 0.499.360.910.780.920.920.9117.0010520CIF
5ABLAPApoN/AN/AN/A N/AN/A0.930.94N/AN/A0.92N/AN/AN/AN/AN/ACIF

Predictions

One row per prediction. Click one to see its pose and interactions, in the panel below.

Run Prediction Protein Ligand Pocket Class Confidence pIC50 Interactions Flags

Overall structure

Interaction diagram

Detected interactions

Metrics

Ligand pose

Superposed targets

Sequence

Pockets

1 named pocket(s) (1N1) plus an unconstrained baseline; contacts are enforced within 4 A.

PocketProteinLigandsRunsTargetsContacts
1N1ABL1IMATI, DASAT2, 4218 residue(s)
UnconstrainedABL1IMATI, DASAT1, 32none -- ligand placed freely

Where the ligands landed

Ligand preparation

2 of 2 ligand(s) flagged for chemistry review -- these are advisory, not errors; verify the input SMILES reflects what you intended before trusting the results below.

Ligand Chemistry notes
IMATI ionizable group(s) present (primary/secondary amine) -- verify the SMILES reflects your intended protonation state
DASAT ionizable group(s) present (primary/secondary amine) -- verify the SMILES reflects your intended protonation state

Ligand structures

No shared scaffold or substructure detected across the set -- ligands are structurally distinct.

Sundefined stereocentreAcarboxylic acidNprimary/secondary aminePhphenolSO3sulfonic acidsaltsalt/disconnected fragment
IMATIN
IMATI structure
MW 494 · cLogP 4.6 · TPSA 86
DASATN
DASAT structure
MW 488 · cLogP 3.3 · TPSA 107

Download PDF · Download SMILES

Scaffolds: Bemis-Murcko, exact match first, then Tanimoto-clustered (Morgan r=2, 2048-bit, threshold 0.60) whole-group MCS as a verified fallback. Minimum highlighted substructure size: 8 heavy atoms. Stereocentre/ionizable-group highlighting from this campaign's own ligand-preparation check (see above).

Interaction counts by type

Click a point to open that target, in the panel below the Targets table.

SSE motif shifts (apo vs holo)

IdentityShiftBoundaryBackboneKinase state
FamilyTargetLigandMotifKindSourceN resRMSD (A)Centroid delta (A)Start deltaEnd deltaFlagged phi/psiDFG apoDFG holoDFG deltaalphaC apoalphaC holoalphaC delta
ABL1ABL1_DASATDASATDFGloopkinase34.473.40N/AN/A2inoutTrueininFalse
ABL1ABL1_DASATDASATDFGloopkinase30.760.67N/AN/A0ininFalseininFalse
ABL1ABL1_DASATDASATcatalytic_looploopkinase30.730.70N/AN/A0inoutTrueininFalse
ABL1ABL1_DASATDASATcatalytic_looploopkinase30.430.41N/AN/A0ininFalseininFalse
ABL1ABL1_DASATDASAThingeloopkinase30.250.12N/AN/A0inoutTrueininFalse
ABL1ABL1_DASATDASAThingeloopkinase30.190.11N/AN/A0ininFalseininFalse
ABL1ABL1_DASATDASATpocket_scaffoldloopkinase731.700.00-146.00-133.008inoutTrueininFalse
ABL1ABL1_DASATDASATpocket_scaffoldloopkinase731.250.00-146.00-133.008ininFalseininFalse
ABL1ABL1_DASATDASATalphaC_Glupointkinase10.970.97N/AN/A0inoutTrueininFalse
ABL1ABL1_DASATDASATalphaC_Glupointkinase10.900.90N/AN/A0ininFalseininFalse
ABL1ABL1_DASATDASATcatalytic_Lyspointkinase11.081.08N/AN/A0inoutTrueininFalse
ABL1ABL1_DASATDASATcatalytic_Lyspointkinase10.880.88N/AN/A0ininFalseininFalse
ABL1ABL1_DASATDASATgatekeeperpointkinase10.090.09N/AN/A0inoutTrueininFalse
ABL1ABL1_DASATDASATgatekeeperpointkinase10.150.15N/AN/A0ininFalseininFalse
ABL1ABL1_IMATIIMATIDFGloopkinase35.584.67N/AN/A2inotherTrueininFalse
ABL1ABL1_IMATIIMATIDFGloopkinase35.234.26N/AN/A2inotherTrueininFalse
ABL1ABL1_IMATIIMATIcatalytic_looploopkinase30.530.49N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIcatalytic_looploopkinase30.650.62N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIhingeloopkinase30.540.43N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIhingeloopkinase30.540.46N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIpocket_scaffoldloopkinase732.030.00-146.00-133.008inotherTrueininFalse
ABL1ABL1_IMATIIMATIpocket_scaffoldloopkinase731.750.00-146.00-133.009inotherTrueininFalse
ABL1ABL1_IMATIIMATIalphaC_Glupointkinase11.511.51N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIalphaC_Glupointkinase11.521.52N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIcatalytic_Lyspointkinase11.411.41N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIcatalytic_Lyspointkinase11.011.01N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIgatekeeperpointkinase10.350.35N/AN/A0inotherTrueininFalse
ABL1ABL1_IMATIIMATIgatekeeperpointkinase10.370.37N/AN/A0inotherTrueininFalse

Download CSV

Family coverage and SSE shifts

FamilyStatusDetail
ABL1OK28 motif row(s) across 4 target(s), annotator=kinase
ABLAPNo apo structure configuredNo 'Apo structure:' configured for this family

Overall shift statistics

  • 2 target(s), 28 motif(s) compared
  • Mean Ca RMSD: 1.32 A (median 0.89 A) — largest shift: 5.58 A at ABL1_IMATI / DFG
  • Mean centroid shift: 0.95 A
  • Flagged phi/psi outlier residues: 39
  • Kinase state changes detected: 21 DFG, 0 alphaC

Per-motif Ca RMSD

Loops

Transmembrane

Selectivity and motif shifts

Family x ligand selectivity

Motif x target RMSD

Ligand pose vs experiment

4 target(s) compared against experimental structures in reference/. Atoms are paired by molecular graph, so the symmetry of the ligand is respected rather than being resolved by whichever atom happened to be nearest.

TargetProteinLigandPocketReferenceSite (A)Pose (A)Conformer (A) 
ABL1_IMATI_1N1ABL1IMATI1N11IEP (STI)0.360.720.29
ABL1_IMATIABL1IMATIunconstrained1IEP (STI)0.381.310.75
ABL1_DASAT_1N1ABL1DASAT1N12GQG (1N1)0.171.851.72
ABL1_DASATABL1DASATunconstrained2GQG (1N1)0.962.221.70

Predicted against experimental

Residue interaction fingerprints

Which residues each ligand touches, per protein. One scale for all of them, so a strong contact looks the same in every plot.

ABL1

The reports as files

The same panels as BoltzMaker wrote them, if you want the whole page in one file.

Landlord narration

5 of 5 target summaries written on-device by the Apple Neural Engine; 0 rendered from the template. Every number here is checked against the figures the analysis computed; a summary stating one it was not given was replaced by the template.

This campaign predicted the binding affinity of five ligands to the ABL1 protein, with three of the five targets deemed to be of high potency, and two of the five targets deemed to be of moderate potency. Of the five predicted targets, two of the targets were flagged as carrying at least one flag, and these were deemed to be of moderate potency.

Targets5 predicted, 4 with a ligand
ReceptorsABL1, ABLAP
LigandsDASAT, IMATI
Verdicts3 proceed, 2 caution
Confidence5 well determined
Flagged2 of 5
Pose validated3 of 4 reproduced the experimental pose

Key findings

  • Of 5 targets, 3 are marked proceed, 2 caution and 0 discard.
  • 5 of 5 targets are well determined; 0 are not.
  • 2 of 5 targets carry at least one flag.
  • 4 targets could be checked against an experimental structure, and 3 reproduced the experimental pose.

Highest predicted potency

  • DASAT on 4_ABL1_DASAT_1N1, predicted pIC50 12.86
  • IMATI on 2_ABL1_IMATI_1N1, predicted pIC50 10.1
  • DASAT on 3_ABL1_DASAT_Unc, predicted pIC50 9.89

Caveats

This campaign failed to flag any of the predicted targets as being of low or no potency. Furthermore, the predicted potency of the ligands may not be applicable to other ligands or conditions.

Per target

TargetVerdictSummaryCaveatWritten by
ABL1_IMATIcautionWell determined; this means that the structure is likely accurate and reliable for the intended use. IMATI is a control ligand with 17 contacts, including 10 hydrophobic contacts, 5 hydrogen bonds, and 2 salt bridges. Its predicted potency is 9.36, which places it at rank 4 out of 4 by predicted potency. The predicted potency varies widely across the ensemble, meaning that the predicted potency of IMATI may not be applicable to other ligands or conditions.The predicted potency varies widely across the ensemble; this means that the predicted potency of IMATI may not be applicable to other ligands or conditions.model
ABL1_IMATI_1N1proceedThe structure is well determined, suggesting it is reliable for further study. IMATI is a control ligand with 16 contacts, 10 hydrophobic and 4 hydrogen bonds, and a predicted potency of 10.1. It is ranked 2 of 4 by predicted potency.No flags were present, indicating no immediate concerns about the structure's validity.model
ABL1_DASATproceedThis structure is well determined, meaning it is likely to be accurate. This is important for scientists who are deciding whether to trust this structure. DASAT is an inhibitor that binds to the target with 11 total contacts (8 hydrophobic and 3 hydrogen bonds). Its predicted potency is 9.89, and it ranks 3 out of 4 by predicted potency.There are no flags indicating any issues with the structure.model
ABL1_DASAT_1N1proceedThis structure is well determined, and that means it should be trusted. DASAT is an inhibitor that binds to ABL1 with 7 hydrophobic contacts and 3 hydrogen bonds, and its predicted potency is 12.86 nM. Its predicted potency ranks it as the most potent inhibitor among four tested ligands.There are no flags indicating problems with the structure. However, the predicted potency of the ligand is not very high, so it may not be a very potent inhibitor.model
ABLAPcautionThis structure is well determined, but caution is still recommended. The ligand is poorly placed relative to the receptor and interaction analysis did not complete for this targetmodel