> ## Documentation Index > Fetch the complete documentation index at: https://docs.biohub.ai/llms.txt > Use this file to discover all available pages before exploring further. # Search for similar proteins using SAE feature vectors > Submit an amino acid sequence and find similar proteins in the atlas based on SAE feature vector similarity. ## OpenAPI ````yaml openapi/metagenomic-atlas-openapi.json GET /esm/protein/api/v1alpha1/similarity-search openapi: 3.1.0 info: title: ESM Atlas API Reference description: Interfaces may change without notice. version: 0.1.0 servers: - url: https://biohub.ai security: [] paths: /esm/protein/api/v1alpha1/similarity-search: get: tags: - Similarity search summary: Search for similar proteins using SAE feature vectors description: >- Submit an amino acid sequence and find similar proteins in the atlas based on SAE feature vector similarity. operationId: sae_search_v2_esm_protein_api_v1alpha1_similarity_search_get parameters: - name: sequence in: query required: true schema: type: string minLength: 1 maxLength: 2048 description: Amino acid sequence title: Sequence description: Amino acid sequence - name: topk_results in: query required: false schema: type: integer maximum: 100 exclusiveMinimum: 0 description: Number of similar proteins to return default: 10 title: Topk Results description: Number of similar proteins to return - name: topk_features in: query required: false schema: type: integer maximum: 100 exclusiveMinimum: 0 description: Number of top features to return default: 20 title: Topk Features description: Number of top features to return - name: min_similarity in: query required: false schema: type: number maximum: 1 minimum: 0 description: >- Minimum similarity score; results below this threshold are excluded default: 0.5 title: Min Similarity description: Minimum similarity score; results below this threshold are excluded - name: cluster_pct_characterized_max in: query required: false schema: anyOf: - type: integer maximum: 100 minimum: 0 - type: 'null' description: >- If set, only return hits whose cluster_pct_characterized is <= this value. Use 0 to find clusters whose members have no characterized Pfam annotations (proxy for uncharacterized / novel). title: Cluster Pct Characterized Max description: >- If set, only return hits whose cluster_pct_characterized is <= this value. Use 0 to find clusters whose members have no characterized Pfam annotations (proxy for uncharacterized / novel). - name: include_cluster_info in: query required: false schema: type: boolean description: >- If true, each result includes the representative protein's cluster_size and human-readable protein_name, sparing the caller a follow-up /clusters/{hash} request per result. default: false title: Include Cluster Info description: >- If true, each result includes the representative protein's cluster_size and human-readable protein_name, sparing the caller a follow-up /clusters/{hash} request per result. responses: '200': description: Successful Response content: application/json: schema: $ref: '#/components/schemas/SAESearchResponseV2' '422': description: Validation Error content: application/json: schema: $ref: '#/components/schemas/HTTPValidationError' components: schemas: SAESearchResponseV2: properties: query_sequence: type: string title: Query Sequence description: Normalized query amino acid sequence protein_hash: anyOf: - type: string - type: 'null' title: Protein Hash description: >- MD5 hash of the query sequence; present only when the protein exists in the atlas similar_proteins: items: $ref: '#/components/schemas/SimilarProtein' type: array title: Similar Proteins description: List of similar proteins top_features_across_results: items: $ref: '#/components/schemas/TopFeatureAcrossResultsV2' type: array title: Top Features Across Results description: Top SAE features commonly activated across similar proteins default: [] restricted_count: type: integer title: Restricted Count description: >- Number of proteins removed from results that are similar to the deny list proteins default: 0 type: object required: - query_sequence - similar_proteins title: SAESearchResponseV2 description: Response schema for V2 SAE search endpoint. HTTPValidationError: properties: detail: items: $ref: '#/components/schemas/ValidationError' type: array title: Detail type: object title: HTTPValidationError SimilarProtein: properties: protein_hash: type: string title: Protein Hash description: MD5 hash of the protein sequence protein_accession: type: string title: Protein Accession description: Source-prefixed protein accession (e.g., 'uniprotkb:P12345') sequence_length: type: integer title: Sequence Length description: Length of the protein sequence similarity_score: type: number title: Similarity Score description: Cosine similarity score (0-1, higher is more similar) pdb: anyOf: - type: string - type: 'null' title: Pdb description: PDB structure string if pre-computed ptm: anyOf: - type: number - type: 'null' title: Ptm description: pTM confidence score mean_plddt: anyOf: - type: number - type: 'null' title: Mean Plddt description: Mean pLDDT confidence score residues_plddt: anyOf: - items: type: number type: array - type: 'null' title: Residues Plddt description: Per-residue pLDDT scores cluster_size: anyOf: - type: integer - type: 'null' title: Cluster Size description: >- Number of proteins in this representative's cluster. Populated only when the request was made with include_cluster_info=true; null otherwise. protein_name: anyOf: - type: string - type: 'null' title: Protein Name description: >- Human-readable cluster name (representative product_name). Populated only when the request was made with include_cluster_info=true. type: object required: - protein_hash - protein_accession - sequence_length - similarity_score title: SimilarProtein description: A similar protein returned by SAE vector search. TopFeatureAcrossResultsV2: properties: feature_index: type: integer title: Feature Index description: SAE feature index (0-16383) occurrence_count: type: integer title: Occurrence Count description: Number of proteins with this feature activated min_activation: type: number title: Min Activation description: Minimum activation value for this feature across result proteins max_activation: type: number title: Max Activation description: Maximum activation value for this feature across result proteins mean_activation: type: number title: Mean Activation description: Mean activation value for this feature across result proteins type: object required: - feature_index - occurrence_count - min_activation - max_activation - mean_activation title: TopFeatureAcrossResultsV2 description: >- Statistics for a feature commonly activated across similar search results. ValidationError: properties: loc: items: anyOf: - type: string - type: integer type: array title: Location msg: type: string title: Message type: type: string title: Error Type input: title: Input ctx: type: object title: Context type: object required: - loc - msg - type title: ValidationError ````