Skip to content

K-Dense-AI/scientific-agent-skills / skills/rowan

Rowan is a cloud-native molecular modeling and medicinal-chemistry workflow platform with a Python API. Use for pKa and macropKa prediction, conformer and tautomer ensembles, docking and analogue docking, protein-ligand cofolding, MSA generation, molecular dynamics, permeability, descriptor workflows, and related small-molecule or protein modeling tasks. Ideal for programmatic batch screening, multi-step chemistry pipelines, and workflows that would otherwise require maintaining local HPC/GPU infrastructure.

references/end_to_end_example.md

View on GitHub
# Example: Tautomer-Aware Analogue Triage

This illustrative campaign preserves molecule objects through geometry workflows,
uses SMILES only for a SMILES pKa method, and records workflow identifiers.
The toy aminopyridines below are not claimed to be active against a target.
Hosted results and docking accuracy were not tested in this refresh.

Requires `rowan-python==3.2.0`, pandas, a configured `ROWAN_API_KEY`, and a
prepared receptor plus validated pocket coordinates for the optional follow-up.

```python
from pathlib import Path
import json
import pandas as pd
import rowan

# Work around the Folder annotation defect in SDK 3.2.0.
from datetime import datetime
rowan.Folder.model_rebuild(_types_namespace={"datetime": datetime})

project = rowan.create_project(name="Analogue triage example")
rowan.project_uuid = project.uuid
folder = rowan.create_folder(name="round_1", parent_uuid=project.root_folder_uuid)
compounds = {"fluoro": "Nc1ncc(F)cc1", "chloro": "Nc1ncc(Cl)cc1", "methoxy": "Nc1ncc(OC)cc1"}

rows = []
selected_tautomers = {}
workflow_ids = {}

def remember(compound, step, workflow):
    workflow_ids.setdefault(compound, {})[step] = workflow.uuid
    Path("campaign_workflows.json").write_text(json.dumps(workflow_ids, indent=2))

for compound, smiles in compounds.items():
    try:
        taut_wf = rowan.submit_tautomer_search_workflow(
            rowan.Molecule.from_smiles(smiles), name=f"{compound} tautomers", folder=folder,
        )
        remember(compound, "tautomers", taut_wf)
        tautomer = taut_wf.result().best_tautomer  # Molecule or None
        if tautomer is None or not tautomer.smiles:
            rows.append({"compound": compound, "error": "No weighted tautomer with SMILES"})
            continue
        selected_tautomers[compound] = tautomer

        pka_wf = rowan.submit_pka_workflow(
            tautomer.smiles, method="starling", name=f"{compound} pKa", folder=folder,
        )
        remember(compound, "pka", pka_wf)
        desc_wf = rowan.submit_descriptors_workflow(
            tautomer, name=f"{compound} descriptors", folder=folder,
        )
        remember(compound, "descriptors", desc_wf)
        pka = pka_wf.result()
        desc = desc_wf.result().descriptors
        if desc is None:
            raise ValueError("Completed descriptor workflow returned no descriptors")
        rows.append({
            "compound": compound,
            "tautomer_smiles": tautomer.smiles,
            "strongest_acid": pka.strongest_acid,
            "strongest_base": pka.strongest_base,
            "exact_mass": desc.get("MW"),
            "topological_psa": desc.get("TopoPSA"),
            "logp": desc.get("SLogP"),
        })
    except rowan.WorkflowError as exc:
        rows.append({"compound": compound, "error": str(exc)})

summary = pd.DataFrame(rows)
summary.to_csv("triage.csv", index=False)
print(summary.to_string(index=False))
```

Do not automatically rank compounds by the smallest pKa: the relevant acid/base
site, desired charge at assay pH, permeability, solubility, and target interactions
all matter. A single highest-weight tautomer is a triage simplification. Carry
multiple populated tautomers/protomers into docking when warranted.

After reviewing the table and experimental objectives, supply `candidate.txt`
with the selected compound identifier, `prepared_receptor.pdb`, and `pocket.json`
containing `[[cx, cy, cz], [sx, sy, sz]]` in Å for that exact receptor:

```python
candidate = Path("candidate.txt").read_text().strip()
if candidate not in selected_tautomers:
    raise ValueError("Selected candidate has no usable tautomer structure")
pocket = json.loads(Path("pocket.json").read_text())
protein = rowan.upload_protein(
    name="prepared target", file_path="prepared_receptor.pdb", project_uuid=project.uuid,
)
docking_wf = rowan.submit_docking_workflow(
    protein=protein,
    pocket=pocket,
    initial_molecule=selected_tautomers[candidate],
    docking_settings=rowan.VinaSettings(scoring_function="vinardo"),
    do_pose_refinement=True,
    name=f"{candidate} docking",
    folder=folder,
)
remember(candidate, "docking", docking_wf)
dock = docking_wf.result()
if not dock.scores:
    raise RuntimeError("No successful docking poses")
print(f"Docking score: {dock.scores[0].score:.2f} kcal/mol")
Path("best_pose.xyz").write_text(dock.best_pose.to_xyz())
if dock.scores[0].complex_pdb:
    dock.get_complex(0).download_structure(name="docked_complex", file_format="mmcif")
```

Set per-workflow `max_credits` to the campaign's agreed limits before execution.
Inspect poses, stereochemistry, clashes, and retained waters/cofactors; validate
with a known ligand where possible. Neither a docking score nor a cofolding
confidence score establishes biological activity. For restartable batches and
submission ambiguity, see [batch_and_webhooks.md](batch_and_webhooks.md).
writing-skillsobra/superpowersUse when creating new skills, editing existing skills, or verifying skills work before deploymentStars: 322kUpdated Claude CodeCodexVerifiedInstall
xlsxanthropics/skillsUse this skill any time a spreadsheet file is the primary input or output. This means any task where the user wants to: open, read, edit, or fix an existing .xlsx, .xlsm, .xltx, .csv, or .tsv file (e.g., adding columns, computing formulas, formatting, charting, cleaning messy data); create a new spreadsheet from scratch or from other data sources; or convert between tabular file formats. Trigger especially when the user references a spreadsheet file by name or path — even casually (like "the xlsx in my downloads") — and wants something done to it or produced from it. Also trigger for cleaning or restructuring messy tabular data files (malformed rows, misplaced headers, junk data) into proper spreadsheets. The deliverable must be a spreadsheet file. Do NOT trigger when the primary deliverable is a Word document, HTML report, standalone Python script, database pipeline, or Google Sheets API integration, even if tabular data is involved.Stars: 198kUpdated Claude CodeCodexVerifiedInstall
shipping-and-launchaddyosmani/agent-skillsPrepares production launches. Use when preparing to deploy to production, or when asking what needs to be in place before shipping. Use when you need a pre-launch checklist, when setting up monitoring, when planning a staged rollout, or when you need a rollback strategy.Stars: 119kUpdated Claude CodeCodexVerifiedInstall
childhoodthedotmack/claude-memBabysit a shipped release after merge, publication, or deployment. Verify published packages, production behavior, installed plugins, and late CI or user reports until the observation window is clean; use babysit for PR readiness before release.Stars: 118kUpdated Claude CodeCodexVerifiedInstall