omics-skills

Review, score, compare, and rank AI-generated biology or bioinformatics research artifacts. Use when auditing AI-scientist notebooks, code, figures, analyses, manuscripts, or reports for rigor, reproducibility, novelty, and task completion.

Search arXiv through its official API and save local Markdown summaries. Use when finding recent CS, math, physics, or quantitative-biology preprints or resolving arXiv IDs.

Create publication-quality static charts with matplotlib or seaborn. Use when scientific figures need readable axes, accessible palettes, tight layouts, and high data-ink design.

Annotate genes or proteins and infer taxonomy from sequence homology. Use when assigning functions, domains, or taxonomic labels to genomes, contigs, or protein sets.

Assemble genomes or metagenomes and assess assembly quality. Use when turning sequence reads into contigs and reporting completeness, continuity, and contamination evidence.

Bin and refine metagenomic contigs, then assess MAG quality. Use when recovering genomes with QuickBin and checking completeness, contamination, and bin consistency.

Curate and validate FASTA or FAA databases. Use when standardizing headers, merging references, deduplicating sequences, converting GenBank files, or preparing BLAST, MMseqs2, and HMM inputs.

Add schema-backed metadata validation, normalized Parquet tables, and a DuckDB catalog to a bioinformatics project. Use when an analysis needs LinkML/Pydantic records or a queryable data catalog.

Call genes and annotate basic sequence features. Use when predicting prokaryotic, viral, or eukaryotic coding sequences before downstream annotation.

Detect and polarize interdomain horizontal gene transfer with homology, context, and phylogenetic checks. Use when studying lateral gene transfer, virus-host gene exchange, endogenous viral elements, or donor direction.

Evaluate scientific claims, methods, biases, and evidence strength. Use when stress-testing a study design, paper, analysis, or causal interpretation.

Build and validate marker-gene alignments and phylogenetic trees. Use when inferring evolutionary relationships, choosing models, or checking tree support and contamination.

Design reproducible bioinformatics pipelines with Prefect plus Dask or Nextflow. Use when scaffolding local, distributed, or scheduler-backed workflows.

Cluster proteins into orthogroups and build pangenome matrices. Use when comparing gene-family presence, absence, expansion, contraction, or core and accessory content across genomes.

Ingest, quality-control, and map sequencing reads with reproducible outputs. Use when processing raw reads, removing contaminants, or calculating mapping and coverage statistics.

Analyze biological results with statistics or machine learning and produce validated reports. Use when aggregating features, testing hypotheses, training models, or reporting performance.

Predict protein structures and perform structure-based annotation. Use when sequence evidence is insufficient or structural similarity, confidence, domains, or complexes matter.

Detect, quality-control, and classify viral contigs. Use when identifying viruses in assemblies, checking viral completeness and contamination, or assigning viral taxonomy.

Generate reproducible Methods from Nextflow, Snakemake, or CWL run artifacts. Use when documenting exact commands, versions, parameters, QC gates, provenance, and outputs.

Structure reproducible bioinformatics projects with canonical layouts, restartable drivers, pinned environments, provenance, and lab notebooks. Use when starting or reorganizing a genomics project or making a sequencing analysis rerunnable.

Search bioRxiv through its official API and filter title, abstract, and author metadata. Use when finding recent biology preprints, scanning date ranges, resolving DOIs, or building author shortlists.

Query Crossref for DOI validation, title matching, citation metadata, and bibliography audits. Use when resolving references or cleaning citation records.

Extract a Conditional Scientific Argumentation Graph and grounded Q&A from a manuscript. Use when representing assertions, contexts, evidence links, and inference steps in machine-readable form.

Inspect scientific data and generate a Markdown structure-and-quality report. Use when triaging tabular, array, sequence, HDF5, JSON, or raster files before downstream analysis.

Run a multi-agent manuscript critique with specialist reports, disagreement checks, and editor synthesis. Use for scientific review, revision assessment, or judging whether an author response resolves prior concerns.

Author, execute, validate, and convert reproducible marimo or Jupyter notebooks. Use when delivering an analysis notebook with all cells run and figures embedded.

Convert PDFs and office documents to clean Markdown, with structured bundles for scientific papers. Use when extracting article structure, preparing a manuscript for analysis, or creating CSAG input.

Build production-ready Plotly Dash dashboards. Use when scientific data needs an interactive, consistently themed layout with clear and performant callbacks.

Search PMC Open Access and bioRxiv corpora with polars-dovmed. Use when structured, reproducible literature queries should run through the hosted API or local parquet indexes.

Produce structured, decision-ready reviews of AI/ML, computational-biology, or bioscience proposals. Use when evaluating grants, projects, or funding applications.

Assess research reach with OpenAlex citations, optional Altmetric data, and journal context. Use when comparing papers, journals, or literature shortlists by influence.

Draft, review, and revise scientific manuscripts with grounded multi-agent checks. Use when writing sections, rebuttals, response letters, manuscript QA, or sentence-level reviews.

Track taxonomy changes and triage sequence assignments across NCBI, GTDB, ICTV, and eukaryotic frameworks. Use when comparing releases, resolving renamed taxa, or routing genomes, bins, and contigs by domain.

Query JGI Lakehouse metadata and retrieve JGI genome or read files. Use when linking GOLD, IMG, MycoCosm, Phytozome, PMO, or JAMO identifiers and datasets.