Brenk filter
Identify toxic and reactive fragments using Brenk structural alert filter.
Related tools
Lead-likeness filter
Screen for lead-like compounds using stricter molecular descriptor criteria than Lipinski or Veber rules for early-stage drug discovery
PAINS filter
Screen compounds for Pan-Assay INterference patterns that cause false positives in biological assays
Veber's rule
Veber's Rule predicts oral bioavailability by evaluating molecular weight, LogP, hydrogen bond donors/acceptors, and rotatable bonds
Lipinski's rule of 5
Lipinski's Rule of Five predicts whether compounds will be orally bioavailable by evaluating molecular weight, LogP, hydrogen bond donors, and acceptors.
ADMET-AI
Predict ADMET (Absorption, Distribution, Metabolism, Excretion, Toxicity) properties from SMILES strings using machine learning models trained on Therapeutics Data Commons datasets.
Admetica
Predict 22 ADMET properties from SMILES strings with the upstream Admetica Chemprop models from Datagrok.
eToxPred
Predict toxicity and synthetic accessibility of small molecules using machine learning. eToxPred combines toxicity risk assessment with synthetic accessibility scoring to help prioritize drug candidates.
QEPPI
Quantitative estimate for protein-protein interaction inhibitor potential. Evaluates drug-likeness for compounds targeting PPIs.
ToxPred 2.0 (Toxicity prediction)
Screen compounds for structural toxicity alerts using PAINS, Brenk, and NIH filters. For focused screening, see PAINS Filter, Brenk Filter, or Veber's Rule.
SPRINT
SPRINT (Structure-aware PRotein ligand INTeraction) predicts drug-target interactions using co-embedded protein and ligand representations. Screen thousands of compounds against a protein target in seconds.
What is the Brenk filter?
The Brenk filter screens compounds for structural features associated with toxicity, chemical reactivity, and poor pharmacokinetics. Published in 2008 by Brenk et al. while assembling screening libraries for neglected disease drug discovery, the filter comprises 105 SMARTS patterns representing problematic molecular substructures.
Unlike property-based rules such as Lipinski or Veber, the Brenk filter identifies specific chemical fragments known to cause problems in drug development. These include mutagenic groups (nitro compounds), reactive functionalities (Michael acceptors, thiols, 2-halopyridines), and features with unfavorable ADMET profiles (sulfates, phosphates, long aliphatic chains).
Categories of structural alerts
The Brenk filter targets several classes of unwanted substructures:
- Mutagenic/toxic groups: Nitro groups, anilines, azo compounds, hydrazines
- Reactive functionalities: Michael acceptors, thiocarbonyls, thiols, halogenated pyridines, acyl halides
- Pharmacokinetic liabilities: Sulfates, phosphates, long aliphatic chains (>C7), crown ethers
- Assay interference: Conjugated nitriles, oxygen-nitrogen single bonds, polyhalogenated compounds
- Unstable moieties: Imines, triple bonds, peroxides, epoxides
The most frequently flagged patterns in typical compound libraries include Michael acceptors, aliphatic long chains, oxygen-nitrogen single bonds, and nitro groups.
How to use the Brenk filter online
ProteinIQ provides browser-based Brenk filtering with immediate results. No software installation required.
Input
| Input | Description |
|---|---|
Molecule | SMILES strings (one per line). Optional compound names can be provided in tab-separated format: name\tSMILES. Accepts file uploads (.smi, .csv, .txt) or PubChem compound IDs. |
Output
| Column | Description |
|---|---|
Name | Compound identifier (auto-generated if not provided) |
SMILES | Input SMILES string |
Brenk alerts | Number of Brenk patterns matched |
Patterns | Names of matched structural alerts (comma-separated) |
Result | Pass if no alerts detected, Fail otherwise |
Results can be exported as CSV, JSON, or Excel for integration with other screening workflows.
Interpreting results
A compound that fails the Brenk filter is not necessarily unsuitable as a drug candidate. The filter serves as an early warning system to prioritize compounds for further investigation.
| Result | Interpretation |
|---|---|
Pass (0 alerts) | No known problematic fragments detected. Proceed with standard evaluation. |
Fail (1-2 alerts) | Contains flagged substructures. Review specific patterns and consider alternatives. |
Fail (3+ alerts) | Multiple structural liabilities. Strong candidates for deprioritization. |
Some marketed drugs contain Brenk-flagged substructures. Context matters: a nitro group in an antibiotic may be acceptable, while the same group in a chronic-use medication warrants concern.
Brenk filter vs. PAINS
The Brenk filter and PAINS filter serve complementary purposes:
| Filter | Focus | Use case |
|---|---|---|
| Brenk | Toxicity, reactivity, pharmacokinetics | Library design, early safety filtering |
| PAINS | Assay interference, promiscuous binding | Hit validation, false positive removal |
The patterns flagged by each filter rarely overlap. Standard practice in virtual screening applies both filters in sequence, though the order depends on the screening stage.
Applications
- Virtual screening: Filtering large compound libraries before docking or machine learning models
- Library curation: Removing unsuitable compounds from corporate or commercial collections
- Lead optimization: Flagging problematic fragments in hit-to-lead chemistry
- Vendor compound selection: Filtering commercial catalogs before procurement
Limitations
The Brenk filter represents patterns associated with problems in drug discovery, not absolute rules:
- Some flagged patterns appear in approved drugs when the benefit-risk profile is acceptable
- The filter cannot predict toxicity magnitude or probability
- Novel problematic substructures discovered after 2008 are not included
- Small molecules with unusual scaffolds may contain unflagged problematic features
For comprehensive safety assessment, combine structural filtering with computational toxicity prediction, ADMET modeling, and experimental testing.