Browse Skills

AlphaFold validates protein designs using AlphaFold2 structure prediction. Use it to predict folding, compute confidence metrics (pLDDT, pTM, ipTM), and validate multi-chain complexes.

AlphaFold Database provides access to over 200 million AI-predicted protein structures from DeepMind and EMBL-EBI. Use it to retrieve predictions by UniProt ID, download PDB/mmCIF files, analyze pLDDT and PAE confidence metrics, and integrate structures into drug discovery or structural biology workflows.

AnnData handles annotated data matrices for single-cell genomics in Python. Use it to read, write, manipulate, and analyze large biological datasets such as h5ad files and single-cell RNA-seq data with metadata.

Antibody Design Agent designs de novo antibodies against viral or tumoral targets using protein language models (MAGE) and RFdiffusion, then validates candidates with AlphaFold3 or ESMFold. Use it when you need binder sequences for a specific antigen or epitope.

Arboreto infers gene regulatory networks from gene expression data using scalable algorithms such as GRNBoost2 and GENIE3. Use it to identify transcription factor–target relationships in bulk or single-cell RNA-seq, with optional distributed computing for large datasets.

Armored CAR-T Design Agent designs armored CAR-T cells with cytokine/chemokine expression for enhanced solid tumor efficacy. It selects armoring payloads like IL-12, IL-15, IL-18, and IL-7, optimizes expression, and integrates safety switches.

arXiv Search searches the complete arXiv database of preprints across physics, mathematics, computer science, and quantitative biology using natural language queries. It's useful for finding preprints before peer review, tracking emerging research trends, or conducting cross-disciplinary searches.

Autonomous Oncology Agent combines LLM reasoning with vision transformer analysis of H&E slides to predict biomarkers, retrieve evidence, and generate ranked treatment options. Use it for tumor board style precision oncology decision support.

BAM Statistics generates alignment statistics from BAM files using samtools and pysam. Use it when assessing alignment quality, calculating coverage, or generating QC reports.

Bayesian Optimizer proposes the next best experiment parameters to maximize a target metric such as yield, purity, or binding affinity, using Gaussian Processes and Upper Confidence Bound acquisition. Use it when experiments are expensive or time-consuming, or to tune machine learning hyperparameters.

Benchling Integration connects agents to Benchling's R&D platform via the Python SDK and REST API for managing DNA/protein entities, inventory, ELN entries, and workflow automation. Use it when automating lab data management or building Benchling Apps.

BGPT Paper Search searches scientific papers and retrieves structured experimental data extracted from full-text studies via the BGPT MCP server. Use it for literature reviews, evidence synthesis, and finding experimental details not available in abstracts alone.