valthos-sdk 0.1.0

Python SDK for Valthos API

Valthos Python SDK

Python SDK for the Valthos bioinformatics platform.

Installation

pip install valthos-sdk

Quick Start

Authentication

First, authenticate with your Valthos account:

import valthos_sdk

# Interactive login (prompts for credentials)
session = valthos_sdk.login()

# Or provide credentials directly
session = valthos_sdk.login(root_domain="valthos.com", token="your-personal-access-token")

Credentials are stored securely in ~/.valthos/credentials.

Working with Workspaces

# List your workspaces
workspaces = session.list_workspaces()
for ws in workspaces:
    print(f"{ws.name}: {ws.status}")

# Get a specific workspace
workspace = session.workspace(name="My Project")

# Create a new workspace
workspace = session.create_workspace(name="New Analysis", description="My analysis project")

File Operations

# Upload a file
workspace.add('/local/path/data.csv', 'data/input.csv')

# List workspace contents
files = workspace.list()
for f in files:
    print(f"{f.path} - {f.size} bytes")

# Create folders
workspace.create_folder('results')

# Copy, rename, delete
workspace.copy('data.csv', 'backup/data.csv')
workspace.rename('old_name.txt', 'new_name.txt')
workspace.delete('temporary_file.txt')

Analysis Tools

Run bioinformatics analyses through the tools interface:

# Fitness analysis
fitness = workspace.tools.Fitness(
    input_path='proteins.parquet',
    output_path='results/fitness.parquet'
)

# Validate inputs
validation = fitness.validate()
if validation['status'] == 'valid':
    # Get runtime estimate
    estimate = fitness.estimate()
    print(f"Estimated time: {estimate['estimated_time_minutes']} minutes")

    # Run the analysis
    result = fitness.run()
    print(f"Job ID: {result['job_id']}")

# Sequence profiling
profiler = workspace.tools.SequenceProfiling(
    input='protein.fasta',
    output='results/profile.json',
    methods=["world"]
)

# Genome profiling
genome_profiler = workspace.tools.GenomeProfiling(
    input='genome.fasta',
    output='results/genome_profile.json',
    min_orf_length=90
)

API Reference

Session Methods

• session.list_workspaces(status=None, limit=None, offset=None) - List workspaces
• session.workspace(name=None, rid=None) - Get a workspace by name or ID
• session.create_workspace(name, description=None) - Create a new workspace
• session.archive_workspace(name=None, rid=None) - Archive a workspace

Workspace Methods

• workspace.add(local_path, workspace_path=None) - Upload a file
• workspace.list(path=None, offset=None, limit=None) - List contents
• workspace.create_folder(path) - Create a folder
• workspace.rename(old_path, new_path) - Rename/move a file or folder
• workspace.copy(source_path, dest_path) - Copy a file
• workspace.delete(path) - Delete a file or folder

Available Tools

• Fitness - Protein fitness inference using ESM2-based models
• SequenceProfiling - Multi-method analysis of protein sequences
• GenomeProfiling - ORF detection and protein profiling from genomes
• SequenceSearch - MMseqs2-based protein similarity search

Error Handling

from valthos import (
    ValthosError,
    AuthenticationError,
    TokenExpiredError,
    APIError,
    WorkspaceNotFoundError,
    FileUploadError
)

try:
    workspace = session.workspace(name="NonExistent")
except WorkspaceNotFoundError:
    print("Workspace not found")
except AuthenticationError:
    print("Please login again")

Requirements

• Python 3.8+
• requests
• pydantic

License

MIT License - see LICENSE for details.