Tools · Computational Biology

Thompson Sampling

Active-learning virtual screening of un-enumerated combinatorial libraries via Thompson Sampling

On this page

Prices, workflows, and method papers Open in the app

Run it from the API

Submit with Submit a job and the job_type below. Price it first with Estimate job reservation cost: submitting reserves that amount from your wallet, and the charge settles at the actual runtime.

Thompson Sampling `thompsonsampling`

Job type: thompsonsampling
Hardware: cpu (default)
Typical runtime: 30 min on CPU

Payload

Provide at least one of: reagents, reagent_lists.

Payload fields
Field	Type	Description
`reaction_smarts`required	`string`	Limits: `min length 1`
`reagents[]`	`reagent_component[]`	Limits: `min items 1`
reagents.`smiles[]`	`string[]`	Limits: `min items 1`
reagents.`file`	`file_payload`
reagents.file.`filename`required	`string`	Limits: `min length 1`
reagents.file.`content`	`string`	Limits: `min length 1`
reagents.file.`content_base64`	`string`	Limits: `min length 1`
`reagent_lists[]`	`string[][]`	Legacy alias for reagents as SMILES lists. Limits: `min items 1`
`num_warmup_trials`required	`integer`	Limits: `≥ 1`
`num_ts_iterations`	`integer`	Default: `2000`Limits: `≥ 1`
`evaluator_class_name`required	`string`	One of: `"FPEvaluator", "MWEvaluator", "MLClassifierEvaluator", "LookupEvaluator"`
`evaluator_arg`	`object`
evaluator_arg.`query_smiles`	`string`	Limits: `min length 1`
evaluator_arg.`ref_colname`	`string`	Limits: `min length 1`
`ts_mode`required	`string`	One of: `"maximize", "minimize", "maximize_boltzmann", "minimize_boltzmann"`
`max_reagents_per_component`	`integer`	Default: `10000`Limits: `≥ 1, ≤ 10000`
`model_file`	`file_payload`
model_file.`filename`required	`string`	Limits: `min length 1`
model_file.`content`	`string`	Limits: `min length 1`
model_file.`content_base64`	`string`	Limits: `min length 1`
`lookup_file`	`file_payload`
lookup_file.`filename`required	`string`	Limits: `min length 1`
lookup_file.`content`	`string`	Limits: `min length 1`
lookup_file.`content_base64`	`string`	Limits: `min length 1`
`known_std`	`number`
`minimum_uncertainty`	`number`

Example

from cognichem_client import CogniChem

client = CogniChem.from_env()  # reads COGNICHEM_API_KEY
payload = {
    "reaction_smarts": "[NH2:2][#6:1].[#6:4][C:3]([OH])=O>>[NH:2]([#6:1])[C:3]([#6:4])=O",
    "reagents": [
        {
            "smiles": [
                "CNC(=O)c1n[nH]c(N)n1",
                "N=C(N)CN1CC[C@H](O)C1",
                "COC[C@@H](O)CN",
                "NC(=O)CN1CCOCC1",
                "CNC(=S)NC(=N)N",
                "… 5 more",
            ],
        },
        {
            "smiles": [
                "N=C(N)NC[C@@H](N)C(=O)O",
                "CN(C)C[C@@H](N)C(=O)O",
                "Nc1nnn(CC(=O)O)n1",
                "COC(=O)[C@@H](O)CC(=O)O",
                "N=C(N)NC[C@H](N)C(=O)O",
                "… 5 more",
            ],
        },
    ],
    "num_warmup_trials": 3,
    "num_ts_iterations": 25,
    "evaluator_class_name": "FPEvaluator",
    "evaluator_arg": {"query_smiles": "CCc1cccc2c(=O)n(C3CNC3)c([C@@H](C)N)nc12"},
    "ts_mode": "maximize",
}

estimate = client.jobs.estimate(job_type="thompsonsampling", payload=payload, resource="cpu")
print(f"Reserves ${estimate.cost:.2f}")

job = client.jobs.submit(
    job_name="my-thompsonsampling-run",
    job_type="thompsonsampling",
    payload=payload,
    resource="cpu",
)
status = client.jobs.wait(job.process_id)
if status.status == "completed":
    client.jobs.result(job.process_id, save_path=".")

curl -X POST "https://api.cognichem.com/api/v1/jobs/submit" \
  -H "X-Api-Key: $COGNICHEM_API_KEY" \
  -H "Idempotency-Key: $(uuidgen)" \
  -H "Content-Type: application/json" \
  -d '{
  "job_name": "my-thompsonsampling-run",
  "job_type": "thompsonsampling",
  "payload": {
    "reaction_smarts": "[NH2:2][#6:1].[#6:4][C:3]([OH])=O>>[NH:2]([#6:1])[C:3]([#6:4])=O",
    "reagents": [
      {
        "smiles": [
          "CNC(=O)c1n[nH]c(N)n1",
          "N=C(N)CN1CC[C@H](O)C1",
          "COC[C@@H](O)CN",
          "NC(=O)CN1CCOCC1",
          "CNC(=S)NC(=N)N",
          "… 5 more"
        ]
      },
      {
        "smiles": [
          "N=C(N)NC[C@@H](N)C(=O)O",
          "CN(C)C[C@@H](N)C(=O)O",
          "Nc1nnn(CC(=O)O)n1",
          "COC(=O)[C@@H](O)CC(=O)O",
          "N=C(N)NC[C@H](N)C(=O)O",
          "… 5 more"
        ]
      }
    ],
    "num_warmup_trials": 3,
    "num_ts_iterations": 25,
    "evaluator_class_name": "FPEvaluator",
    "evaluator_arg": {
      "query_smiles": "CCc1cccc2c(=O)n(C3CNC3)c([C@@H](C)N)nc12"
    },
    "ts_mode": "maximize"
  },
  "resource": "cpu"
}'

{
  "reaction_smarts": "[NH2:2][#6:1].[#6:4][C:3]([OH])=O>>[NH:2]([#6:1])[C:3]([#6:4])=O",
  "reagents": [
    {
      "smiles": [
        "CNC(=O)c1n[nH]c(N)n1",
        "N=C(N)CN1CC[C@H](O)C1",
        "COC[C@@H](O)CN",
        "NC(=O)CN1CCOCC1",
        "CNC(=S)NC(=N)N",
        "… 5 more"
      ]
    },
    {
      "smiles": [
        "N=C(N)NC[C@@H](N)C(=O)O",
        "CN(C)C[C@@H](N)C(=O)O",
        "Nc1nnn(CC(=O)O)n1",
        "COC(=O)[C@@H](O)CC(=O)O",
        "N=C(N)NC[C@H](N)C(=O)O",
        "… 5 more"
      ]
    }
  ],
  "num_warmup_trials": 3,
  "num_ts_iterations": 25,
  "evaluator_class_name": "FPEvaluator",
  "evaluator_arg": {
    "query_smiles": "CCc1cccc2c(=O)n(C3CNC3)c([C@@H](C)N)nc12"
  },
  "ts_mode": "maximize"
}

Sample data from the job catalog; long values are shortened here. Each job_name must be unique among your jobs.

Workflow inputs

None

Workflow outputs

ArchiveZIP
MoleculesSMILES

Run it from the API

Thompson Sampling thompsonsampling

Thompson Sampling `thompsonsampling`