---
title: RDKit Quickstart
description: 'Paste SMILES one per line, pick a preset — drug-like filter, similarity search, diverse subset, 3D conformers — choose operations and submit.'
---

You want to compute molecular descriptors, fingerprints, or drug-likeness flags for a set of SMILES. This runs it from the webapp: pick a preset, paste your SMILES, choose operations, and submit — no code required.

## Webapp input methods

The .bloom* webapp provides a comprehensive interface for RDKit cheminformatics analysis.

## Quick presets

Click any preset to auto-configure operations:

| Preset | Operations | Use Case |
|--------|------------|----------|
| **Drug-like Filter** | Descriptors + Filter | Lipinski/PAINS screening |
| **Similarity Search** | Fingerprints + Similarity | Find analogs |
| **Scaffold Analysis** | Murcko Scaffold + Descriptors | Core structure extraction |
| **Diverse Subset** | Diverse Pick + Descriptors | Library selection |
| **3D Conformers** | Conformers + Descriptors | Docking prep |
| **Comprehensive** | Descriptors + Fingerprints + Filter | Full analysis |

## SMILES input

- **Format**: One SMILES per line
- **Placeholder**: Examples show ethanol (`CCO`) and aspirin
- **Validation**: Real-time SMILES parsing and error display

## Operation selection

### Core operations (chips)

| Operation | Description |
|-----------|-------------|
| **Descriptors** | Calculate molecular property descriptors |
| **Fingerprints** | Generate molecular fingerprints |
| **3D Conformers** | Generate 3D conformations |
| **Filter** | Apply drug-likeness filters |
| **Similarity** | Calculate pairwise similarity |

### Advanced operations (expandable)

| Operation | Description |
|-----------|-------------|
| **Murcko Scaffold** | Extract core scaffolds |
| **Diverse Pick** | Select maximally diverse subset |
| **Substructure Search** | Find SMARTS patterns |
| **MCS** | Maximum common substructure |
| **Standardize** | Canonicalize structures |
| **Fingerprint Bits** | Explain fingerprint contributions |
| **RascalMCES** | Multi-fragment MCS analysis |

## Parameter tabs

### Descriptors tab

- **Descriptor Set**: Drug-like (13), Lipinski (4), Physicochemical, Fragments (85), Topological (19), All (200+)

### Fingerprints tab

| Parameter | Options/Range |
|-----------|---------------|
| **Type** | Morgan, RDKit, MACCS, Atom Pair, Topological Torsion |
| **Radius** | 1-6 (default: 2) |
| **Bits** | 128-16384 (default: 2048) |

### 3D Conformers tab

- **Number of Conformers**: 1-100 (default: 10)
- **Force Field**: MMFF94 or UFF

### Similarity tab

- **Reference SMILES**: Required for similarity search
- **Metric**: Tanimoto, Dice, Cosine
- **Threshold**: 0-1 (default: 0.3)

### Filters tab

**Rule-based**: Lipinski Rule of 5, Veber Rules

**Substructure (SMARTS)**: PAINS, Brenk, NIH filters

**Property ranges**: MW Min/Max

### Export tab

- **Export CSV**: Enabled by default
- **Export SDF**: For 3D coordinates

## Submit button

- **Label**: `Run RDKit ({n} operations)`
- **Shows operation count dynamically

---

## Cheminformatics analysis with RDKit in .bloom* platform

## Step-by-step molecular analysis workflow

### 1. Access RDKit analysis tools

Open **RDKit Full Analysis** in the **Data & Literature** section of .bloom*'s Tools panel, or ask Ivy in chat. RDKit provides comprehensive molecular analysis capabilities including structure validation, property calculation, and molecular manipulation directly through the web interface.

### 2. Prepare your molecular data

**Input format options:**
- **SMILES strings**: Direct input of chemical structures
- **SDF files**: Upload multi-molecule structure files with 3D coordinates
- **CSV files**: Tabular data with SMILES in the first column
- **Previous jobs**: Import molecules from other .bloom* tools and analyses

**Data validation:**
- Automatic SMILES parsing and structure validation
- Format detection and conversion between representations
- Molecular weight and composition verification
- Stereochemistry and charge state assessment

### 3. Choose analysis workflow

**Structure analysis:**
- **Molecular properties**: Calculate physicochemical descriptors
- **Structure validation**: Check for chemical validity and common issues
- **Substructure search**: Identify molecular patterns and functional groups
- **Similarity calculation**: Compare molecules using various metrics

**Conformational analysis:**
- **3D structure generation**: Create low-energy conformations
- **Conformer ensembles**: Generate multiple molecular shapes
- **Energy minimization**: Optimize molecular geometries
- **Conformational analysis**: Study molecular flexibility

**Molecular descriptors:**
- **Topological descriptors**: Graph-based molecular properties
- **Electrostatic properties**: Charge distributions and potentials
- **Pharmacophore features**: Identify key interaction points
- **Fingerprint generation**: Create molecular fingerprints for similarity

### 4. Configure analysis parameters

**Basic settings:**
- Select appropriate analysis type for your research question
- Choose output formats (tabular data, molecular files, visualizations)
- Set quality thresholds and filtering criteria
- Configure computational parameters for accuracy vs. speed

**Advanced options:**
- **Conformer generation**: Number of conformers and energy thresholds
- **Fingerprint types**: Choose appropriate fingerprint methods
- **Similarity metrics**: Select comparison algorithms (Tanimoto, Dice, etc.)
- **Property calculations**: Customize descriptor sets

### 5. Execute analysis and explore results

**Real-time processing:**
- Progress tracking through molecular analysis phases
- Interactive preview of results as they become available
- Error handling with detailed feedback for problematic molecules
- Batch processing with parallel computation

**Interactive results exploration:**
- **Molecular viewer**: 2D and 3D structure visualization
- **Property tables**: Sortable results with statistical summaries
- **Similarity matrices**: Heatmap visualization of molecular relationships
- **Distribution plots**: Histograms and scatter plots of molecular properties

**Analysis tools:**
- **Filtering and sorting**: Interactive data exploration
- **Statistical analysis**: Distribution analysis and outlier detection
- **Comparative visualization**: Side-by-side molecular comparisons
- **Export capabilities**: Multiple formats for downstream analysis

### 6. Export and integrate results

**Export formats:**
- **Property tables**: CSV/TSV files with calculated descriptors
- **Molecular files**: SDF/PDB formats with coordinates and properties
- **Fingerprint data**: Binary or count fingerprints for machine learning
- **Visualization exports**: Images and interactive molecular views

**Integration options:**
- **Direct to predictive tools**: Send molecules to ADMET-AI or DeepSol
- **Molecular docking**: Use structures with DiffDock
- **Generative chemistry**: Feed results to DeepPurpose or AiZynthFinder
- **Database export**: Save results for external analysis tools

## Common cheminformatics analysis scenarios

### Compound library characterization
1. Upload compound collection in SMILES or SDF format
2. Calculate comprehensive molecular descriptors
3. Analyze property distributions across the library
4. Identify outliers and assess library quality
5. Generate property-based filtering criteria

### Structure-property relationship studies
1. Import compound series with activity data
2. Calculate molecular descriptors and fingerprints
3. Perform similarity analysis within series
4. Identify structural features correlated with activity
5. Generate SAR analysis reports

### Molecular diversity assessment
1. Upload compound library for analysis
2. Generate molecular fingerprints
3. Calculate pairwise similarity matrix
4. Assess diversity using clustering and PCA
5. Identify representative compounds for screening

### Conformation analysis for docking
1. Upload small molecule structures
2. Generate multiple conformers using RDKit
3. Energy minimize and filter low-energy conformations
4. Export conformation ensemble for docking studies
5. Prepare structures for DiffDock analysis

## Quality control and validation

### Input validation checklist

**Molecular structure:**
- [ ] Valid SMILES or SDF format
- [ ] Chemically reasonable structures (no unusual valences)
- [ ] Appropriate molecular weights and sizes
- [ ] Correct stereochemistry specifications

**Data integrity:**
- [ ] Consistent formatting across input files
- [ ] No duplicate molecules unless intentional
- [ ] Complete data for all required fields
- [ ] Appropriate file sizes for analysis scope

### Analysis validation

**Computational results:**
- [ ] Property calculations within expected ranges
- [ ] Structure generation produces valid 3D coordinates
- [ ] Fingerprint generation completes without errors
- [ ] Similarity calculations are numerically reasonable

**Biological relevance:**
- [ ] Calculated properties align with known values
- [ ] Structural features match intended chemistry
- [ ] Analysis results support research hypotheses
- [ ] Quality metrics meet project requirements

## Integration with .bloom* workflows

### Upstream data sources

**Compound libraries:**
- **ZINC database**: Direct integration with compound searches
- **PubChem data**: Import from chemical database queries
- **Generative chemistry**: Analyze molecules from DeepPurpose/AiZynthFinder
- **Experimental data**: Process compounds from screening campaigns

**Research data:**
- **SAR data**: Structure-activity relationship datasets
- **Property data**: Experimental measurements for validation
- **Spectral data**: NMR/IR data for structure confirmation
- **Biological data**: Activity data for correlation analysis

### Downstream applications

**Predictive modeling:**
- **ADMET predictions**: Use calculated properties as model inputs
- **Activity modeling**: Generate descriptors for QSAR studies
- **Property prediction**: Validate calculated vs. measured properties
- **Virtual screening**: Prepare molecular fingerprints

**Molecular design:**
- **Similarity searching**: Find analogs in compound databases
- **Scaffold analysis**: Identify common structural frameworks
- **Fragment identification**: Break molecules into building blocks
- **Synthetic planning**: Assess molecular complexity

## Performance optimization

### Computational efficiency

**Batch processing:**
- Process multiple molecules simultaneously
- Utilize parallel computation for large datasets
- Optimize memory usage for different analysis types
- Monitor computational resources and timing

**Algorithm selection:**
- Choose appropriate methods for dataset size and complexity
- Balance accuracy with computational requirements
- Use fast approximations for initial screening
- Apply rigorous methods for final analysis

### Scalability considerations

**Large dataset handling:**
- Implement chunked processing for massive libraries
- Use streaming algorithms for memory efficiency
- Apply filtering early to reduce computational load
- Consider distributed processing for very large analyses

**Result management:**
- Implement result caching for repeated analyses
- Use compressed formats for large output files
- Configure appropriate precision for numerical results
- Plan data storage and retrieval strategies

## Best practices for cheminformatics analysis

### Data preparation

**Input quality:**
- Validate molecular structures before analysis
- Standardize chemical representations (canonical SMILES)
- Handle salts, counterions, and mixtures appropriately
- Document data sources and processing steps

**Analysis planning:**
- Choose appropriate methods for research objectives
- Consider computational requirements and time constraints
- Plan for result interpretation and validation
- Document analysis parameters for reproducibility

### Result interpretation

**Chemical context:**
- Interpret results in context of known chemistry
- Consider molecular size and complexity effects
- Account for calculation method limitations
- Validate key results with experimental data

**Statistical considerations:**
- Understand descriptor calculation methods and ranges
- Consider normalization and scaling for comparisons
- Account for dataset biases and sampling effects
- Use appropriate statistical tests for significance

### Research reproducibility

**Documentation:**
- Record all analysis parameters and settings
- Preserve input data and processing scripts
- Document software versions and calculation methods
- Generate comprehensive analysis reports

**Collaboration:**
- Share analysis methods and results with team members
- Generate publication-ready figures and tables
- Document workflows for peer review
- Maintain searchable result archives

## Troubleshooting common issues

### Structure parsing problems
- Check SMILES syntax and atom specifications
- Verify stereochemistry notation
- Handle special characters and formatting issues
- Use structure validation tools to identify problems

### Computational errors
- Check molecular size limits for different algorithms
- Verify input data completeness and consistency
- Monitor memory usage for large datasets
- Review error messages for specific guidance

### Result interpretation challenges
- Cross-validate calculated properties with literature values
- Consider calculation method appropriateness for molecule type
- Account for solvent and condition effects on properties
- Validate against experimental measurements when available

### Performance issues
- Optimize batch sizes for available computational resources
- Use appropriate algorithms for dataset characteristics
- Implement result filtering to reduce output volume
- Consider alternative methods for problematic molecules

## Read next