Esmc Tools#

Configuration File: remote_tools/esmc_tools.json Tool Type: Remote Tools Count: 2

This page contains all tools defined in the esmc_tools.json configuration file.

Available Tools#

run_esmc_embed (Type: RemoteTool)#

Embed one protein sequence with the ESM Cambrian (ESM C) protein language model and return the me…

run_esmc_embed tool specification

Tool Information:

  • Name: run_esmc_embed

  • Type: RemoteTool

  • Description: Embed one protein sequence with the ESM Cambrian (ESM C) protein language model and return the mean-pooled embedding vector. Use it to compare, cluster or search proteins by learned representation. The full per-residue matrix is NOT returned inline (2560 floats per residue): ask for ‘residue_range’ to get a bounded window of residue vectors, or ‘per_residue_file’ to get the whole matrix written as a .npy file whose path comes back.

Parameters:

  • sequence (string) (required) One protein sequence in single-letter amino-acid code, or a single-record FASTA. At most 4096 residues.

  • include_cls_embedding (boolean) (optional) Also return the BOS (sequence-level) vector, which is another whole embedding of the same width.

  • residue_range (array) (optional) Inclusive 1-based [start, end] window of residue vectors to return inline. Refused, not clipped, when (end - start + 1) * embedding_dim exceeds 20000 floats – at the 6B model’s 2560 dimensions that is 7 residues.

  • per_residue_file (boolean) (optional) Write the whole (length x embedding_dim) float32 matrix to a .npy file on the provider host and return its path, shape and size instead of its contents.

  • decimals (integer) (optional) Decimal places for returned floats (default 5).

Example Usage:

query = {
    "name": "run_esmc_embed",
    "arguments": {
        "sequence": "example_value"
    }
}
result = tu.run(query)

run_esmc_logits (Type: RemoteTool)#

Score one protein sequence with the ESM Cambrian (ESM C) masked language model: the model’s mean …

run_esmc_logits tool specification

Tool Information:

  • Name: run_esmc_logits

  • Type: RemoteTool

  • Description: Score one protein sequence with the ESM Cambrian (ESM C) masked language model: the model’s mean log-likelihood and pseudo-perplexity for the sequence, plus the residues it finds least likely. Use it to rank designs, flag unusual positions, or compare variants of one protein. Per-position log-probabilities are opt-in and bounded, because a long sequence is a long array.

Parameters:

  • sequence (string) (required) One protein sequence in single-letter amino-acid code, or a single-record FASTA. At most 4096 residues.

  • top_n (integer) (optional) How many least-likely residues to name (default 10).

  • per_residue_log_prob (boolean) (optional) Return the log-probability of every observed residue, one number per position. Refused for sequences longer than 4000 residues.

  • decimals (integer) (optional) Decimal places for returned floats (default 5).

Example Usage:

query = {
    "name": "run_esmc_logits",
    "arguments": {
        "sequence": "example_value"
    }
}
result = tu.run(query)