EvolutionaryScale · Schema

EvolutionaryScale ESMProtein

Schema for the `ESMProtein` object exposed by the EvolutionaryScale `esm` Python SDK and accepted by the Forge ESM3 API. An ESMProtein is the canonical multi-track representation of a protein covering sequence, structure, secondary structure, SASA, and function annotations.

AIArtificial IntelligenceBiologyBioinformaticsComputational BiologyDrug DiscoveryESMESM3ESM CambrianFoundation ModelsGenerative BiologyLife SciencesMachine LearningProtein DesignProtein FoldingProtein Language ModelsProteinsRepresentation LearningStructure Prediction

Properties

Name Type Description
sequence string Amino acid sequence in one-letter codes (A, R, N, D, C, E, Q, G, H, I, L, K, M, F, P, S, T, W, Y, V). An underscore `_` denotes a masked position to be filled by generation.
coordinates array Per-residue atom37 coordinates, shape [L, 37, 3]. Missing atoms are represented as nulls or NaNs depending on serialization.
secondary_structure string DSSP-style per-residue secondary structure string. Common codes: H (alpha-helix), E (beta-strand), C (coil).
sasa array Per-residue solvent accessible surface area (square Angstroms).
function_annotations array Optional list of per-residue function annotations referencing InterPro / GO / Pfam labels.
plddt array Per-residue pLDDT confidence score (0.0–100.0).
ptm number Predicted TM-score (0.0–1.0) for the structure.
View JSON Schema on GitHub

JSON Schema

evolutionaryscale-esm-protein-schema.json Raw ↑ 
{
  "$schema": "https://json-schema.org/draft/2020-12/schema",
  "$id": "https://api-evangelist.com/schemas/evolutionaryscale/evolutionaryscale-esm-protein-schema.json",
  "title": "EvolutionaryScale ESMProtein",
  "description": "Schema for the `ESMProtein` object exposed by the EvolutionaryScale `esm` Python SDK and accepted by the Forge ESM3 API. An ESMProtein is the canonical multi-track representation of a protein covering sequence, structure, secondary structure, SASA, and function annotations.",
  "type": "object",
  "properties": {
    "sequence": {
      "type": "string",
      "description": "Amino acid sequence in one-letter codes (A, R, N, D, C, E, Q, G, H, I, L, K, M, F, P, S, T, W, Y, V). An underscore `_` denotes a masked position to be filled by generation.",
      "pattern": "^[ARNDCEQGHILKMFPSTWYV_X*-]+$",
      "example": "MKTAYIAKQRQISFVK_____SSERVKKLLVGDIVT"
    },
    "coordinates": {
      "type": "array",
      "description": "Per-residue atom37 coordinates, shape [L, 37, 3]. Missing atoms are represented as nulls or NaNs depending on serialization.",
      "items": {
        "type": "array",
        "minItems": 37,
        "maxItems": 37,
        "items": {
          "type": "array",
          "minItems": 3,
          "maxItems": 3,
          "items": { "type": "number" }
        }
      }
    },
    "secondary_structure": {
      "type": "string",
      "description": "DSSP-style per-residue secondary structure string. Common codes: H (alpha-helix), E (beta-strand), C (coil)."
    },
    "sasa": {
      "type": "array",
      "description": "Per-residue solvent accessible surface area (square Angstroms).",
      "items": { "type": "number", "minimum": 0 }
    },
    "function_annotations": {
      "type": "array",
      "description": "Optional list of per-residue function annotations referencing InterPro / GO / Pfam labels.",
      "items": {
        "type": "object",
        "properties": {
          "label": { "type": "string" },
          "start": { "type": "integer", "minimum": 1 },
          "end": { "type": "integer", "minimum": 1 },
          "source": { "type": "string" }
        }
      }
    },
    "plddt": {
      "type": "array",
      "description": "Per-residue pLDDT confidence score (0.0–100.0).",
      "items": { "type": "number", "minimum": 0, "maximum": 100 }
    },
    "ptm": {
      "type": "number",
      "description": "Predicted TM-score (0.0–1.0) for the structure.",
      "minimum": 0,
      "maximum": 1
    }
  },
  "anyOf": [
    { "required": ["sequence"] },
    { "required": ["coordinates"] }
  ],
  "additionalProperties": false
}