> ## Documentation Index > Fetch the complete documentation index at: https://proto.evodesign.org/docs/llms.txt > Use this file to discover all available pages before exploring further. # Overall Protein Quality > Evaluate overall protein quality using multiple sub-constraints
Overall Protein Quality
**License:** This constraint can use multiple tools, each under its own license. See the **Tools Used** tab and each tool's page for license details.

This constraint is open source. Any third-party models, product names, or trademarks referenced are the property of their respective owners, and Proto is not affiliated with them.

evo-design/proto-language/proto\_language/constraint/protein\_quality/overall\_protein\_quality\_constraint.py
View source
Constraint contributors
Evaluate overall protein quality using multiple configurable sub-constraints. This constraint function provides a comprehensive assessment of protein quality by evaluating multiple aspects including sequence length, structural complexity, repetitiveness, amino acid diversity, and balanced amino acid representation. For DNA sequences, it first predicts protein-coding regions using Prodigal, then evaluates all predicted proteins. For protein sequences, it evaluates them directly. The function aggregates scores from enabled sub-constraints by averaging them and clipping to \[0.0, 1.0]. Use the native `Constraint(threshold=...)` parameter for pass/fail filtering. ## API Reference
ConfigOverallProteinQualityConfig Source
Configuration for the overall protein quality constraint. This configuration class orchestrates multiple protein quality sub-constraints that can be enabled or disabled individually. It provides a flexible framework for comprehensive protein quality assessment by combining various metrics including sequence length, structural complexity, repetitiveness, amino acid diversity, and balanced amino acid representation. The configuration uses a nested structure where all sub-constraint parameters are exposed through a single `protein_quality_config` attribute of type `ProteinQualitySubConfig`. This design allows for easy serialization in UI/API schemas while maintaining clear organization of constraint-specific parameters. At least one sub-constraint must be enabled for the configuration to be valid. This is enforced through a model validator that runs after initialization. The nested `protein_quality_config` provides access to: * **Length constraint**: Validates protein length against min/max range or target value * **Complexity constraint**: Detects low-complexity regions using segmasker * **Repetitiveness constraint**: Identifies repeated k-mer patterns * **Diversity constraint**: Ensures adequate amino acid type diversity * **Balanced amino acids constraint**: Checks for underrepresented amino acid types Each sub-constraint can be independently enabled/disabled and configured with specific parameters. See `ProteinQualitySubConfig` documentation for complete parameter details. For more details, see: * `ProteinQualitySubConfig`: Detailed documentation of all sub-constraint parameters and configuration options * `overall_protein_quality_constraint`: The constraint function that uses this configuration * `SequenceLengthConfig`: Configuration for length constraint * `ProteinComplexityConfig`: Configuration for complexity constraint * `ProteinRepetitivenessConfig`: Configuration for repetitiveness constraint * `ProteinDiversityConfig`: Configuration for diversity constraint * `BalancedAaConfig`: Configuration for balanced amino acids constraint Nested configuration for protein quality checks
ReturnsConstraintOutput
One result per sequence. Scores range from 0.0 (best) to 1.0 (worst) and represent the average of all enabled sub-constraint scores, clipped to \[0.0, 1.0]. For DNA sequences, the score reflects the average quality across all predicted proteins. `metadata` carries: **For DNA sequences:** * `prodigal_proteins`: List of dicts of predicted proteins from Prodigal, each with protein ID, sequence, length, etc. (or `None` if no ORFs were predicted) * `prodigal_protein_count`: Integer count of predicted ORFs * `predicted_protein_count`: Integer count of proteins (same as prodigal\_protein\_count) * `avg_constraint_score`: Float average quality score across all predicted proteins * `protein_quality_details`: List of dictionaries, one per predicted protein, each containing: * `protein_id`: String identifier from Prodigal * `length`: Integer protein length in amino acids * `avg_constraint_score`: Float average across enabled constraints * `quality_scores`: Dictionary mapping constraint names to scores * `metadata`: Dictionary of additional constraint-specific metadata **For protein sequences:** * `protein_quality_scores`: Dictionary mapping constraint names (e.g., "length", "complexity", "repetitiveness", "diversity", "balanced\_aas") to their individual scores * `avg_constraint_score`: Float average across all enabled constraints
## Usage Using all available constraints with custom thresholds: ```python python icon="python" theme={null} >>> quality_config = ProteinQualitySubConfig( ... enable_length=True, ... length_target_length=300, ... enable_complexity=True, ... complexity_max_low_complexity=0.25, ... enable_repetitiveness=True, ... repetitiveness_max_repetitiveness=0.08, ... repetitiveness_min_repeat_length=3, ... enable_diversity=True, ... diversity_min_diversity=0.75, ... enable_balanced_aas=True, ... balanced_min_aa_frequency=0.03, ... balanced_max_underrepresented_count=2, ... ) >>> overall_cfg = OverallProteinQualityConfig(protein_quality_config=quality_config) >>> protein_seq = Sequence("MKYIVAVAG...", "protein") >>> results = overall_protein_quality_constraint([(protein_seq,)], overall_cfg) ``` ## Metadata | Property | Value | | --------------- | ------------------------------------ | | Key | `overall-protein-quality` | | Function | `overall_protein_quality_constraint` | | Category | `protein_quality` | | Mode | `discrete` | | Uses GPU | `False` | | Supported Types | `dna`, `protein` |