> ## 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.
# Protein Domain Match
> Evaluate whether sequences contains protein domains matching specified keywords
**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.
Evaluate whether sequences contain protein domains matching specified keywords.
This constraint function searches for functional protein domains using HMMER's
hmmscan tool against HMM profile databases. It identifies domains in protein
sequences and matches them against user-specified keywords, enabling selection
of proteins with desired functional domains.
For DNA sequences, the function first runs Prodigal to predict protein-coding
regions (ORFs), then searches each predicted protein for matching domains. For
protein sequences, the domain search is performed directly. The constraint is
satisfied when the specified keyword criteria are met (any or all keywords,
depending on configuration).
## API Reference
Configuration for protein domain constraint.
This class defines configuration parameters for evaluating whether protein
sequences contain specific functional domains identified by keyword searches
against HMM (Hidden Markov Model) profile databases. The constraint uses
HMMER's hmmscan tool to identify protein domains and matches them against
user-specified keywords, enabling targeted selection for proteins with
desired functional characteristics.
For DNA sequences, Prodigal is used to predict ORFs first, then each
predicted protein is searched for domains. For protein sequences, the
search is performed directly.
Path to HMM database file for hmmscan (e.g., Pfam-A.hmm). Must be pressed with hmmpress.
Keywords to search for in domain descriptions (case-insensitive).
Maximum E-value for significant HMM hits; lower is more stringent (typical range 0.0001 to 0.01).
Min query coverage percentage for significant hits (0-100).
If True, require ALL keywords to be found. If False, require ANY keyword (default).
Configuration for PyHMMER hmmscan.
ReturnsConstraintOutput
One result per sequence. A score of 0.0 indicates
domain criteria are satisfied (matching domains found) and 1.0 indicates
no matching domains found or failure to meet keyword requirements.
`metadata` carries:
**For DNA sequences:**
* `prodigal_proteins`: List of dicts of predicted proteins from
Prodigal (or `None` if no ORFs were predicted)
* `prodigal_protein_count`: Integer count of predicted ORFs
* `domain_search_results`: List of domain search results for each
predicted protein
* `domain_keywords_found`: List of unique keywords found across all
predicted proteins
* `domain_matching_proteins`: List of protein IDs that matched keywords
**For protein sequences:**
* `domain_search_results`: List containing domain search results
* `domain_keywords_found`: List of keywords found in domain descriptions
* `domain_matching_hits`: DataFrame of domain hits matching keywords
* `hmmscan_all_hits`: DataFrame of all significant hmmscan hits
## Usage
Evaluating domain presence in protein with single keyword:
```python python icon="python" theme={null}
>>> from proto_language.core import Sequence, SequenceType
>>> seq = Sequence("MVLSPADKTNVKAAWGKVGAHAGEYGAEALERMFLSF", "protein")
>>> cfg = ProteinDomainConfig(hmm_db="Pfam-A.hmm", keywords=["kinase"], evalue_threshold=0.001)
>>> results = protein_domain_constraint([(seq,)], config=cfg)
>>> print(results[0].score) # 0.0 if kinase domain found, 1.0 if not
>>> print(results[0].metadata["domain_keywords_found"]) # ['kinase'] if found
```
Evaluating DNA sequence (with automatic ORF prediction):
```python python icon="python" theme={null}
>>> dna_seq = Sequence("ATGGTACTGAGCCCAGCG...", "dna")
>>> cfg = ProteinDomainConfig(hmm_db="Pfam-A.hmm", keywords=["helicase"])
>>> results = protein_domain_constraint([(dna_seq,)], config=cfg)
>>> print(results[0].metadata["prodigal_protein_count"]) # Number of predicted ORFs
>>> print(results[0].metadata["domain_matching_proteins"]) # IDs of proteins with helicase domain
```
## Metadata
| Property | Value |
| --------------- | --------------------------- |
| Key | `protein-domain` |
| Function | `protein_domain_constraint` |
| Category | `protein_quality` |
| Mode | `discrete` |
| Uses GPU | `False` |
| Supported Types | `dna`, `protein` |
evo-design/proto-language/proto\_language/constraint/protein\_quality/protein\_domain\_constraint.py
