Skip to main content

Available Tools

Computational bioinformatics tools available in SmartsBio — AI protein design, sequence alignment, quality control, variant analysis, file processing, and more. Tools can be called directly via the API or invoked automatically by the agent.

Looking for database tools? Sequence search, protein databases, genomics, pathways, clinical data, literature, and patents are covered on the Available Databases page.
Sync

Results returned immediately via POST /v1/tools/:id/run. Best for fast, lightweight operations.

Pipeline

Async execution via POST /v1/pipelines. Get a pipeline ID immediately, then stream or poll. Used for compute-heavy tools.

Both

Supports either mode. A single long-running tool submitted to /v1/pipelines is valid — no multi-step chain required.

Tool Reference

32 tools across 8 categories. Tool IDs are stable across versions.

AI Protein Design6 tools

Tool IDNameDescriptionMode
protein_structure_predictionProtein Structure PredictionAI-powered 3D protein structure prediction using Boltz-2. Accepts sequence or FASTA, outputs PDB/mmCIF.Pipeline
binding_affinity_predictionBinding Affinity PredictionPredict binding affinity between a protein binder and target using Boltz-2 thermodynamic scoring on GPU.Pipeline
protein_binder_designProtein Binder DesignDesign protein binders with enhanced binding affinity for a given target structure.Pipeline
peptide_binder_designPeptide Binder DesignDesign short peptide binders for target proteins, optimizing for binding specificity.Pipeline
nanobody_designNanobody DesignDesign nanobodies (VHH antibodies) with high specificity for a target epitope.Pipeline
enzyme_designEnzyme DesignDesign enzymes for targeted catalytic functions using generative protein models.Pipeline

Sequence Alignment3 tools

Tool IDNameDescriptionMode
bwa_toolkitBWABurrows-Wheeler Aligner for short read alignment to reference genomes. Outputs SAM/BAM.Pipeline
bowtie2_toolkitBowtie2Ultrafast and memory-efficient short read aligner for DNA and RNA-seq reads.Pipeline
hisat2_toolkitHISAT2Hierarchical, splice-aware aligner for RNA-seq reads. Supports spliced and alternative alignment.Pipeline

Quality Control3 tools

Tool IDNameDescriptionMode
fastqc_toolkitFastQCQuality assessment of raw high-throughput sequencing data. Generates per-base quality plots.Pipeline
trimmomatic_toolkitTrimmomaticRead trimming and adapter removal for paired-end and single-end Illumina data.Pipeline
fmlrc_toolkitFMLRCFM-index based long-read error correction using short-read data as reference.Pipeline

Variant Analysis3 tools

Tool IDNameDescriptionMode
gatk_toolkitGATKGATK4 variant discovery and genotyping on AWS infrastructure. Germline and somatic calling.Pipeline
annovar_toolkitANNOVARComprehensive variant annotation with multiple databases (refGene, COSMIC, ClinVar, gnomAD). Requires license.Pipeline
vcftools_toolkitVCFtoolsVCF/BCF file filtering, statistics, format conversion, and population genetics analyses.Both

Genomic File Processing7 tools

Tool IDNameDescriptionMode
bedtools_toolkitBEDToolsGenome interval manipulation: intersect, merge, complement, closest, coverage for BED/GFF/VCF.Both
picard_toolkitPicardHigh-throughput sequencing data manipulation: mark duplicates, sort, index, and compute metrics.Pipeline
tabix_queryTabixFast indexed random-access queries on bgzipped genomic files (VCF, BED, GFF).Sync
genomicranges_toolkitGenomicRangesBioconductor GenomicRanges for interval operations, overlap analysis, and range arithmetic.Sync
format_conversion_toolkitFormat ConversionConvert between bioinformatics formats: FASTA/FASTQ, SAM/BAM, VCF/BCF, GFF/GTF, BED, and archives.Both
sequence_editor_toolkitSequence EditorDNA/RNA/protein sequence editing: trim, translate, reverse complement, GC content, motif search.Sync
zip_toolkitArchive / ZipFile compression and decompression: ZIP, GZIP, BZIP2, XZ, 7-Zip, and TAR archive formats.Sync

Analysis & Statistics2 tools

Tool IDNameDescriptionMode
data_statisticsData StatisticsDescriptive statistics, t-test, ANOVA, correlation analysis, and outlier detection on tabular data.Sync
visualization_generatorVisualization GeneratorCreate charts, heatmaps, volcano plots, Manhattan plots, and 3D structure visualizations.Sync

File Management3 tools

Tool IDNameDescriptionMode
file_readerFile ReaderRead and parse workspace files: Word, Excel, CSV, PDF (with OCR via AWS Textract), FASTA, GenBank.Sync
file_writerFile WriterSave reports, tables, and documents to the workspace.Sync
file_searchFile SearchFind and discover files in the workspace by name, extension, or metadata.Sync

Pipeline Management5 tools

Tool IDNameDescriptionMode
start_pipelineStart PipelineLaunch a predefined multi-step bioinformatics pipeline by pipeline ID or custom steps.Both
pipeline_statusPipeline StatusCheck progress of a running multi-step pipeline — current step, logs, and completion percentage.Sync
list_pipelinesList PipelinesDiscover predefined bioinformatics pipelines with descriptions, steps, and estimated runtimes.Sync
process_statusProcess StatusCheck status of a long-running background process by process ID.Sync
process_resultsProcess ResultsRetrieve output files and results from a completed background process.Sync
Looking for predefined pipelines? Multi-step bioinformatics workflows (WES alignment, RNA-seq, GATK variant calling, protein binder design, and more) are documented on the Predefined Pipelines page.

Fetch the Tool List at Runtime

Call GET /v1/tools to retrieve the live tool registry with each tool's ID, parameter schema, and description.

from smartsbio import SmartsBio

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

tools = client.tools.list()
for tool in tools:
    print(f"{tool.id:40s} [{tool.category}]")

# Filter by category
protein_design = [t for t in tools if t.category == "AI Protein Design"]
pipeline_tools = [t for t in tools if t.invocation in ("pipeline", "both")]

# Inspect a tool's parameter schema
blast = next(t for t in tools if t.id == "ncbi_blast")
for param in blast.parameters:
    print(f"  {param.name} ({'required' if param.required else 'optional'}): {param.description}")

Run a Tool Directly

# Sync tool — results returned immediately
result = client.tools.run(
    tool_id="ncbi_pubmed",
    input={"query": "BRCA1 breast cancer 2024", "maxResults": 5},
)
for article in result["results"]:
    print(f"{article['pmid']}: {article['title']}")

# Pipeline tool — submit and poll
pipeline = client.pipelines.create(
    tool_id="protein_structure_prediction",
    workspace_id=ws_id,
    input={"sequence": "MKTIIALSYIFCLVFA...", "model": "boltz2"},
)
pipeline = client.pipelines.wait(pipeline["id"], workspace_id=ws_id,
                                  on_progress=lambda p: print(p["progress_pct"], "%"))
client.files.download(pipeline["output_keys"][0], workspace_id=ws_id, dest="./structure.pdb")