The Bio Research Stack

Modern biology research is increasingly computational. Single-cell sequencing, high-dimensional phenotyping, and large-scale omics datasets mean that even bench scientists need to write pipelines, process instrument output, and select problems with a clear sense of what the data infrastructure will support. Claude skills for life sciences researchers and bioinformaticians cover this engineering-adjacent work — from framing the scientific problem at the start of a project through QC, analysis tooling, pipeline development, and instrument data standardization.

This stack was built for researchers who need structured, reproducible support for the computational side of modern biology. Every skill follows a consistent workflow: define inputs, validate conditions, execute steps in order, produce documented output.

What these skills do

Bio Research Start

Structure the beginning of a research project — defining the scientific question, mapping the experimental approach, identifying the key dependencies and failure modes, and the project framing that keeps complex multi-step research programs on track from day one. Use it at the start of any new project to establish the scope and deliverables before work begins.

Bio Research Scientific Problem Selection

Evaluate and select research problems — novelty assessment, feasibility analysis, impact framing, and the problem selection process that helps researchers invest in questions that are both tractable and scientifically meaningful. Covers literature-informed framing and gap identification. Particularly useful when choosing between competing directions at the start of a grant cycle or rotation.

Bio Research Single Cell RNA QC

Run quality control on single-cell RNA sequencing data — cell filtering, doublet detection, ambient RNA removal, mitochondrial fraction assessment, and the QC workflow that ensures downstream analysis is built on clean, reliable data. Covers standard scRNA-seq pipelines with Scanpy, Seurat, and related tools.

Bio Research scVI Tools

Apply scVI (single-cell Variational Inference) tools to single-cell data — latent space modeling, batch correction, differential expression, and the scVI workflow that handles large-scale single-cell datasets with probabilistic modeling. Covers the scverse ecosystem including scVI-tools, scANVI, totalVI, and related packages.

Bio Research Nextflow Development

Develop and maintain Nextflow bioinformatics pipelines — workflow structure, process definitions, channel logic, executor configuration, and the pipeline engineering that makes reproducible, scalable analyses possible across compute environments including HPC clusters, AWS Batch, and Google Cloud Life Sciences.

Bio Research Instrument Data to Allotrope

Convert instrument data to the Allotrope Data Format (ADF) — schema mapping, metadata extraction, file conversion, and the standardization workflow that makes instrument output interoperable with downstream analysis and regulatory submission requirements. Covers mass spectrometry, chromatography, plate reader, and other common lab instrument formats.

How the stack works together

Starting a project: Use Bio Research Start to define the scope and experimental design, then Bio Research Scientific Problem Selection to validate that the question is novel, feasible, and impactful before committing resources.

Running a single-cell analysis: Use Bio Research Single Cell RNA QC to clean the data first — filter low-quality cells, remove doublets, assess ambient RNA contamination. Once QC passes, use Bio Research scVI Tools for batch correction, latent space representation, and differential expression analysis.

Building reproducible pipelines: Use Bio Research Nextflow Development to wrap QC and analysis steps into portable, reproducible workflows that run consistently across local development, HPC, and cloud environments.

Standardizing instrument output: Use Bio Research Instrument Data to Allotrope to convert raw instrument files to ADF before they enter the analysis pipeline — ensuring metadata is captured and the data is in a format that meets regulatory and interoperability requirements.

Who this is for

• Computational biologists and bioinformaticians building and running analysis pipelines
• Life sciences researchers working with single-cell sequencing, spatial transcriptomics, or other high-dimensional omics data
• Research engineers at pharma, biotech, and academic labs developing reproducible bioinformatics workflows
• Graduate students and postdocs who need structured workflow support for complex multi-step analyses
• Data scientists at CROs and research hospitals handling instrument data standardization and regulatory submissions

Installing this stack

Each skill installs independently with a single command from the skill’s page. Drop the extracted folder into ~/.claude/skills/ and it’s active in your next Claude Code session — no configuration required. Skills activate automatically when you describe a task that matches their trigger conditions.

For broader data science and statistical analysis tools, see the Data Science & Analytics Stack. For compliance documentation in regulated life sciences contexts, see the Compliance & Quality Stack.

→ View all research skills · Full install guide