Rowan: Cloud-Native Molecular-Modeling and Drug-Design Workflows

Overview

Rowan is a cloud-native workflow platform for molecular simulation, medicinal chemistry, and structure-based design. Its Python API exposes a unified interface for small-molecule modeling, property prediction, docking, molecular dynamics, and AI structure workflows.

Use Rowan when you want to run medicinal-chemistry or molecular-design workflows programmatically without maintaining local HPC infrastructure, GPU provisioning, or a collection of separate modeling tools. Rowan handles all infrastructure, result management, and computation scaling.

When to use Rowan

Rowan is a good fit for:

• Quantum chemistry, semiempirical methods, or neural network potentials
• Batch property prediction (pKa, descriptors, permeability, solubility)
• Conformer and tautomer ensemble generation
• Docking workflows (single-ligand, analogue series, pose refinement)
• Protein-ligand cofolding and MSA generation
• Multi-step chemistry pipelines (e.g., tautomer search → docking → pose analysis)
• Batch medicinal-chemistry campaigns where you need consistent, scalable infrastructure

Rowan is not the right fit for:

• Simple molecular I/O (use RDKit directly)
• Post-HF ab initio quantum chemistry or relativistic calculations

Access and pricing model

Rowan uses a credit-based usage model. All users, including free-tier users, can create API keys and use the Python API.

Free-tier access

• Access to all Rowan core workflows
• 20 credits per week
• 500 signup credits

Pricing and credit consumption

Credits are consumed according to compute type:

• CPU: 1 credit per minute
• GPU: 3 credits per minute
• H100/H200 GPU: 7 credits per minute

Purchased credits are priced per credit and remain valid for up to one year from purchase.

Typical cost estimates

Quick start

uv pip install rowan-python

import rowan
rowan.api_key = "your_api_key_here"  # or set ROWAN_API_KEY env var

# Submit a descriptors workflow — completes in under a minute
wf = rowan.submit_descriptors_workflow("CC(=O)Oc1ccccc1C(=O)O", name="aspirin")
result = wf.result()

print(result.descriptors['MW'])    # 180.16
print(result.descriptors['SLogP']) # 1.19
print(result.descriptors['TPSA'])  # 59.44

If that prints without error, you're set up correctly.

Installation

uv pip install rowan-python
# or: pip install rowan-python

User and webhook management

Authentication

Set an API key via environment variable (recommended):

export ROWAN_API_KEY="your_api_key_here"

Or set directly in Python:

import rowan
rowan.api_key = "your_api_key_here"

Verify authentication:

import rowan
user = rowan.whoami()  # Returns user info if authenticated
print(f"User: {user.email}")
print(f"Credits available: {user.credits_available_string}")

Webhook secret management

For webhook signature verification, manage secrets through your user account:

import rowan

# Get your current webhook secret (returns None if none exists)
secret = rowan.get_webhook_secret()
if secret is None:
    secret = rowan.create_webhook_secret()
print(f"Secret key: {secret.secret}")

# Rotate your secret (invalidates old, creates new)
# Use this periodically for security
new_secret = rowan.rotate_webhook_secret()
print(f"New secret created (old secret disabled): {new_secret.secret}")

# Verify incoming webhook signatures
is_valid = rowan.verify_webhook_secret(
    request_body=b"...",           # Raw request body (bytes)
    signature="X-Rowan-Signature", # From request header
    secret=secret.secret
)

Molecule input formats

Rowan accepts molecules in the following formats:

• SMILES (preferred): "CCO", "c1ccccc1O"
• SMARTS patterns (for some workflows): subset of SMARTS for substructure matching
• InChI (if supported in your API version): "InChI=1S/C2H6O/c1-2-3/h3H,2H2,1H3"

The API will validate input and raise a rowan.ValidationError if a molecule cannot be parsed. Always use canonicalized SMILES for reproducibility.

Tip: Use RDKit to validate SMILES before submission:

from rdkit import Chem
smiles = "CCO"
mol = Chem.MolFromSmiles(smiles)
if mol is None:
    raise ValueError(f"Invalid SMILES: {smiles}")

Core usage pattern

Most Rowan tasks follow the same three-step pattern:

1. Submit a workflow
2. Wait for completion (with optional streaming)
3. Retrieve typed results with convenience properties

import rowan

# 1. Submit — use the specific workflow function (not the generic submit_workflow)
workflow = rowan.submit_descriptors_workflow(
    "CC(=O)Oc1ccccc1C(=O)O",
    name="aspirin descriptors",
)

# 2. & 3. Wait and retrieve
result = workflow.result()  # Blocks until done (default: wait=True, poll_interval=5)
print(result.data)              # Raw dict
print(result.descriptors['MW']) # 180.16 — use result.descriptors dict, not result.molecular_weight

For long-running workflows, use streaming:

for partial in workflow.stream_result(poll_interval=5):
    print(f"Progress: {partial.complete}%")
    print(partial.data)

result() vs. stream_result()

Guideline: Use result() for descriptors, pKa. Use stream_result() for conformer search, docking, cofolding.

Working with results

Rowan's API includes typed workflow result objects with convenience properties.

Using typed properties and .data

Results have two access patterns:

1. Convenience properties (recommended first): result.descriptors, result.best_pose, result.conformer_energies
2. Raw fallback: result.data — raw dictionary from the API

Example:

result = rowan.submit_descriptors_workflow(
    "CCO",
    name="ethanol",
).result()

# Convenience property (returns dict of all descriptors):
print(result.descriptors['MW'])   # 46.042
print(result.descriptors['SLogP'])  # -0.001
print(result.descriptors['TPSA'])   # 57.96

# Raw data fallback (descriptors are nested under 'descriptors' key):
print(result.data['descriptors'])
# {'MW': 46.042, 'SLogP': -0.001, 'TPSA': 57.96, 'nHBDon': 1.0, 'nHBAcc': 1.0, ...}

Note: DescriptorsResult does not have a molecular_weight property. Descriptor keys use short names (MW, SLogP, nHBDon) not verbose names.

Cache invalidation

Some result properties are lazily loaded (e.g., conformer geometries, protein structures). To refresh:

result.clear_cache()
new_structures = result.conformer_molecules  # Refetched

Projects, folders, and organization

For nontrivial campaigns, use projects and folders to keep work organized.

Projects

import rowan

# Create a project
project = rowan.create_project(name="CDK2 lead optimization")
rowan.set_project("CDK2 lead optimization")

# All subsequent workflows go into this project
wf = rowan.submit_descriptors_workflow("CCO", name="test compound")

# Retrieve later
project = rowan.retrieve_project("CDK2 lead optimization")
workflows = rowan.list_workflows(project=project, size=50)

Folders

# Create a hierarchical folder structure
folder = rowan.create_folder(name="docking/batch_1/screening")

wf = rowan.submit_docking_workflow(
    # ... docking params ...
    folder=folder,
    name="compound_001",
)

# List workflows in a folder
results = rowan.list_workflows(folder=folder)

Workflow decision trees

pKa vs. MacropKa

Use microscopic pKa when:

• You need the pKa of a single ionizable group
• You're interested in acid–base transitions and protonation thermodynamics
• The molecule has one or two ionizable sites
• Speed is critical (faster, fewer credits)

Use macropKa when:

• You need pH-dependent behavior across a physiologically relevant range (e.g., 0–14)
• You want aggregated charge and protonation-state populations across pH
• The molecule has multiple ionizable groups with coupled protonation
• You need downstream properties like aqueous solubility at different pH

Example decision:

Phenol (pKa ~10): Use microscopic pKa
Amine (pKa ~9–10): Use microscopic pKa
Multi-ionizable drug (N, O, acidic group): Use macropKa
ADME assessment across GI pH: Use macropKa

Conformer search vs. tautomer search

Use conformer search when:

• A single tautomeric form is known
• You need a diverse 3D ensemble for docking, MD, or SAR analysis
• Rotatable bonds dominate the chemical space

Use tautomer search when:

• Tautomeric equilibrium is uncertain (e.g., heterocycles, keto–enol systems)
• You need to model all relevant protonation isomers
• Downstream calculations (docking, pKa) depend on tautomeric form

Combined workflow:

# Step 1: Find best tautomer
taut_wf = rowan.submit_tautomer_search_workflow(
    initial_molecule="O=c1[nH]ccnc1",
    name="imidazole tautomers",
)
best_taut = taut_wf.result().best_tautomer

# Step 2: Generate conformers from best tautomer
conf_wf = rowan.submit_conformer_search_workflow(
    initial_molecule=best_taut,
    name="imidazole conformers",
)

Docking vs. analogue docking vs. cofolding

Common workflow categories

1. Descriptors

A lightweight entry point for batch triage, SAR, or exploratory scripts.

wf = rowan.submit_descriptors_workflow(
    "CC(=O)Oc1ccccc1C(=O)O",  # positional arg, accepts SMILES string
    name="aspirin descriptors",
)

result = wf.result()
print(result.descriptors['MW'])    # 180.16
print(result.descriptors['SLogP']) # 1.19
print(result.descriptors['TPSA'])  # 59.44
print(result.data['descriptors'])
# {'MW': 180.16, 'SLogP': 1.19, 'TPSA': 59.44, 'nHBDon': 1.0, 'nHBAcc': 4.0, ...}

Common descriptor keys:

The result contains hundreds of additional molecular descriptors (BCUT, GETAWAY, WHIM, etc.); access any via result.descriptors['key'].

2. Microscopic pKa

For protonation-state energetics and acid/base behavior of a specific structure.

Two methods are available:

# Fast path: SMILES input with full acid+base coverage (use starling method when available)
wf = rowan.submit_pka_workflow(
    initial_molecule="c1ccccc1O",       # phenol SMILES; param is initial_molecule, not initial_smiles
    method="starling",   # fast SMILES method, covers acid+base; chemprop_nevolianis2025 is deprotonation-only
    name="phenol pKa",
)

result = wf.result()
print(result.strongest_acid)    # 9.81 (pKa of the most acidic site)
print(result.conjugate_bases)   # list of {pka, smiles, atom_index, ...} per deprotonatable site

3. MacropKa

For pH-dependent protonation behavior across a range.

wf = rowan.submit_macropka_workflow(
    initial_smiles="CN1CCN(CC1)C2=NC=NC3=CC=CC=C32",  # imidazole
    min_pH=0,
    max_pH=14,
    min_charge=-2,  # default
    max_charge=2,   # default
    compute_aqueous_solubility=True,  # default
    name="imidazole macropKa",
)

result = wf.result()
print(result.pka_values)               # list of pKa values
print(result.logd_by_ph)               # dict of {pH: logD}
print(result.aqueous_solubility_by_ph) # dict of {pH: solubility}
print(result.isoelectric_point)        # isoelectric point
print(result.data)
# {'pKa_values': [...], 'logD_by_pH': {...}, 'aqueous_solubility_by_pH': {...}, ...}

4. Conformer search

For 3D ensemble generation when ensemble quality matters.

wf = rowan.submit_conformer_search_workflow(
    initial_molecule="CCOC(=O)N1CCC(CC1)Oc1ncnc2ccccc12",
    num_conformers=50,  # Optional: override default
    name="conformer search",
)

result = wf.result()
print(result.conformer_energies)  # [0.0, 1.2, 2.5, ...]
print(result.conformer_molecules)  # List of 3D molecules
print(result.best_conformer)  # Lowest-energy conformer

5. Tautomer search

For heterocycles and systems where tautomer state affects downstream modeling.

wf = rowan.submit_tautomer_search_workflow(
    initial_molecule="O=c1[nH]ccnc1",  # or keto tautomer
    name="imidazolone tautomers",
)

result = wf.result()
print(result.best_tautomer)  # Most stable SMILES string
print(result.tautomers)      # List of tautomeric SMILES
print(result.molecules)      # List of molecule objects

6. Docking

For protein-ligand docking with optional pose refinement and conformer generation.

# Upload protein once, reuse in multiple workflows
protein = rowan.upload_protein(
    name="CDK2",
    file_path="cdk2.pdb",
)

# Define binding pocket
pocket = {
    "center": [10.5, 24.2, 31.8],
    "size": [18.0, 18.0, 18.0],
}

# Submit docking
wf = rowan.submit_docking_workflow(
    protein=protein,
    pocket=pocket,
    initial_molecule="CCNc1ncc(c(Nc2ccc(F)cc2)n1)-c1cccnc1",
    do_pose_refinement=True,
    do_conformer_search=True,
    name="lead docking",
)

result = wf.result()
print(result.scores)  # Docking scores (kcal/mol)
print(result.best_pose)  # Mol object with 3D coordinates
print(result.data)  # Raw result dict

Protein preparation tips:

• PDB files should be reasonably clean (remove water/heteroatoms unless intended)
• Use the same protein object across a docking series for consistency
• If you have a PDB ID, use rowan.create_protein_from_pdb_id() instead

7. Analogue docking

For placing a compound series into a shared binding context.

# Analogue series (e.g., SAR campaign)
analogues = [
    "CCNc1ncc(c(Nc2ccc(F)cc2)n1)-c1cccnc1",    # reference
    "CCNc1ncc(c(Nc2ccc(Cl)cc2)n1)-c1cccnc1",   # chloro
    "CCNc1ncc(c(Nc2ccc(OC)cc2)n1)-c1cccnc1",   # methoxy
    "CCNc1ncc(c(Nc2cc(C)c(F)cc2)n1)-c1cccnc1", # methyl, fluoro
]

wf = rowan.submit_analogue_docking_workflow(
    analogues=analogues,
    initial_molecule=analogues[0],  # Reference ligand
    protein=protein,
    pocket=pocket,
    name="SAR series docking",
)

result = wf.result()
print(result.analogue_scores)  # List of scores for each analogue
print(result.best_poses)  # List of poses

8. MSA generation

For multiple-sequence alignment (useful for downstream cofolding).

wf = rowan.submit_msa_workflow(
    initial_protein_sequences=[
        "MENFQKVEKIGEGTYGVVYKARNKLTGEVVALKKIRLDTETEGVP"
    ],
    output_formats=["colabfold", "chai", "boltz"],
    name="target MSA",
)

result = wf.result()
result.download_files()  # Downloads alignments to disk

9. Protein-ligand cofolding

For AI-based bound-complex prediction when no crystal structure is available.

wf = rowan.submit_protein_cofolding_workflow(
    initial_protein_sequences=[
        "MENFQKVEKIGEGTYGVVYKARNKLTGEVVALKKIRLDTETEGVP"
    ],
    initial_smiles_list=[
        "CCNc1ncc(c(Nc2ccc(F)cc2)n1)-c1cccnc1"
    ],
    name="protein-ligand cofolding",
)

result = wf.result()
print(result.predictions)  # List of predicted structures
print(result.messages)  # Model metadata/warnings

predicted_structure = result.get_predicted_structure()
predicted_structure.write("predicted_complex.pdb")

All supported workflow types

All workflows follow the same submit → wait → retrieve pattern and support webhooks and project/folder organization.

Core molecular modeling workflows

Structure-based design workflows

Advanced computational chemistry

Reaction chemistry

Advanced properties

Binding free energy

Sequence and structural biology

Batch submission and retrieval

For libraries or analogue series, submit in a loop using the specific workflow function. The generic rowan.batch_submit_workflow() and rowan.submit_workflow() functions currently return 422 errors from the API — use the named functions (submit_descriptors_workflow, submit_pka_workflow, etc.) instead.

Submit a batch

smileses = ["CCO", "CC(=O)O", "c1ccccc1O"]
names = ["ethanol", "acetic acid", "phenol"]

workflows = [
    rowan.submit_descriptors_workflow(smi, name=name)
    for smi, name in zip(smileses, names)
]

print(f"Submitted {len(workflows)} workflows")

Poll batch status

statuses = rowan.batch_poll_status([wf.uuid for wf in workflows])
# Returns aggregate counts — not per-UUID:
# {'queued': 0, 'running': 1, 'complete': 2, 'failed': 0, 'total': 3, ...}

if statuses["complete"] == statuses["total"]:
    print("All workflows done")
elif statuses["failed"] > 0:
    print(f"{statuses['failed']} workflows failed")

Retrieve and collect results

results