@basemark/bio

Bioinformatics components for Markdown. Write a short identifier — an accession, a PDB ID, an rsID, a locus — and get back a real, interactive viewer: protein structures, sequence tracks, clinical variants, pathway maps, interaction networks, genome browsing, and more.

Install

bun add @basemark/core @basemark/bio

Usage

import { createRegistry, renderMarkdown } from '@basemark/core';
import { registerBioComponents } from '@basemark/bio';

const registry = createRegistry();
await registerBioComponents(registry);

renderMarkdown('::structure{pdbid="1cbs"}', registry);

In React, use @basemark/react's MarkdownRenderer instead of calling renderMarkdown directly (still await registration first):

import { MarkdownRenderer } from '@basemark/react';

<MarkdownRenderer source={'::structure{pdbid="1cbs"}'} registry={registry} />;

Each component below is a directive you write directly in Markdown, e.g. ::protvista{accession="P05067"}.

Components

Protein structure & sequence

• structure — 3D protein structure from a PDB entry (3Dmol.js)
• protvista — UniProt sequence, domain, and feature tracks (ProtVista)

Variants & clinical genomics

• variant — a ClinVar/dbSNP variant card from an rsID: gene, position, allele, clinical significance, associated conditions, and a CADD score

Genomic loci & association plots

• locuszoom-assoc — regional GWAS association plot
• locuszoom-gwas-catalog — association plot with published GWAS Catalog hits labeled
• locuszoom-phewas — phenome-wide association scan for one variant
• locuszoom-intervals — association plot with interval/regulatory annotations
• locuszoom-credible-sets — 95% Bayesian credible set for fine-mapping
• locuszoom-multi-pheno — several related phenotypes overlaid on one locus
• genome-browser — scrollable, zoomable genome browser for a locus (IGV.js)

Pathways & networks

• pathway — a KEGG pathway diagram from a pathway ID
• interaction-network — a protein-protein interaction network from one or more gene names (STRING)

Sequences & phylogenetics

• fasta — a sequence viewer with a position ruler, residue coloring, and highlightable ranges
• newick — a phylogenetic tree from Newick-format text

Coming soon

• gene — Ensembl gene track
• AlphaFold predicted structures, for proteins without a solved PDB entry
• Karyotype/ideogram — chromosome-level view with cytobands
• Taxonomy/species tree
• RNA secondary structure
• Sequence logo / motif viewer
• Plasmid / circular map viewer
• Pedigree chart
• Multiple sequence alignment (MSA) viewer
• Whole-slide/microscopy imaging

locuszoom-tabix (tabix-indexed files) is implemented but not registered by default — a known vendored-dependency issue under strict ESM. Call registerTabix() directly to opt in.

Chemistry components (SMILES, PubChem, reaction schemes) are a separate package — see packages/chem (not started yet).