smarts-bio 0.1.2

Official Python SDK for smarts.bio

Run BLAST, GATK, AlphaFold, BWA, DESeq2, and dozens more bioinformatics tools — all through a single API.

---

Get started in 2 minutes

1. Sign up at smarts.bio and create a free account
2. Generate an API key — Organization Settings → API Keys
3. Install and run:

pip install smarts-bio

from smartsbio import SmartsBio

client = SmartsBio(api_key="sk_live_...")

response = client.query.run(
    "Run BLAST for ATGCGTAACCGTAA and find homologs in nr database",
    workspace_id="ws_abc",
)
print(response.answer)

Full documentation → smarts.bio/docs

---

What can you do?

Ask the AI agent anything — it orchestrates the right tools automatically:

Category	Tools
Sequence analysis	BLAST, HMMER, Clustal Omega, MUSCLE
Variant calling	GATK HaplotypeCaller, FreeBayes, DeepVariant
Alignment	BWA-MEM, STAR, HISAT2, Bowtie2
Structure prediction	AlphaFold, RoseTTAFold, ESMFold
RNA-seq / expression	DESeq2, edgeR, Salmon, kallisto
Genome annotation	Prokka, Augustus, BRAKER
Literature & databases	PubMed, NCBI Gene, UniProt, STRING, ClinVar
Pipelines	WES alignment, somatic variant calling, RNA-seq differential expression

These are just some of the tools available. See the full updated list at smarts.bio/docs.

---

Installation

pip install smarts-bio

Requirements: Python 3.9+. Single dependency: httpx.

---

Authentication

# Pass directly
client = SmartsBio(api_key="sk_live_...")

# Or set SMARTSBIO_API_KEY environment variable
client = SmartsBio()

Generate your key at chat.smarts.bio → Organization Settings → API Keys.

---

Examples

Ask a bioinformatics question

response = client.query.run(
    "Find BRCA1 variants associated with breast cancer and summarize the evidence",
    workspace_id="ws_abc",
)
print(response.answer)

Real-time streaming

for chunk in client.query.stream("Align these reads to GRCh38", workspace_id="ws_abc"):
    if chunk.type == "status":
        print(f"[{chunk.status}]")
    elif chunk.type == "content":
        print(chunk.content, end="", flush=True)
    elif chunk.type == "done":
        print("\nDone.")

Run a bioinformatics pipeline

# Upload your FASTQ files
r1 = client.files.upload("sample_R1.fastq.gz", "ws_abc")
r2 = client.files.upload("sample_R2.fastq.gz", "ws_abc")

# Launch a WES alignment pipeline
pipeline = client.pipelines.create(
    "ws_abc",
    input={"fastq_r1": r1.key, "fastq_r2": r2.key, "reference": "GRCh38"},
    pipeline_id="alignment-wes",
)

# Wait for results (polls automatically)
result = client.pipelines.wait(
    pipeline.id,
    "ws_abc",
    on_progress=lambda p: print(f"  {p.progress_pct}% — {p.current_step}"),
)

Async client

import asyncio
from smartsbio import SmartsBioAsync

async def main():
    async with SmartsBioAsync(api_key="sk_live_...") as client:
        response = await client.query.run(
            "Find BRCA1 variants",
            workspace_id="ws_abc",
        )
        print(response.answer)

asyncio.run(main())

Visualize results

result = client.visualizations.viewer_url("orgs/.../variants.vcf", "ws_abc")
print(f"Open in browser: {result.viewer_url}")

---

SDK Modules

Module	Description
client.query	Ask the AI agent — sync or streaming SSE
client.workspaces	List and manage workspaces
client.conversations	Retrieve conversation history
client.tools	List available tools and run them directly
client.files	Upload, download, and manage files
client.pipelines	Launch and monitor long-running bioinformatics pipelines
client.visualizations	Generate shareable viewer URLs and render plots

---

Error Handling

from smartsbio import AuthenticationError, PermissionDeniedError, RateLimitError, APIError

try:
    client.query.run("...")
except AuthenticationError:
    print("Invalid API key")
except PermissionDeniedError as e:
    print(f"Key lacks required scope: {e.required_scope}")
except RateLimitError as e:
    print(f"Rate limited — retry after {e.retry_after}s")
except APIError as e:
    print(f"API error {e.status}: {e.message}")

---

Configuration

client = SmartsBio(
    api_key="sk_live_...",
    timeout=120,       # seconds (default: 120)
    max_retries=3,     # retries on 429 / 5xx (default: 3)
)

---

More Examples

See the examples/ directory:

• query_sync.py — basic AI query
• query_stream.py — real-time streaming output
• query_async.py — async client with streaming
• list_tools.py — enumerate available tools
• upload_and_pipeline.py — upload files and run a pipeline
• viewer_url.py — generate a bio-viewer link

---

Documentation & Support

• Full docs: smarts.bio/docs
• Platform: smarts.bio
• Issues: GitHub Issues
• Email: support@smarts.bio

---

Built with ❤️ by the smarts.bio team