tahoe-explorer
Filter and subset Tahoe-100M perturbation metadata, 100.6 million rows, and leave with a reproducible pull recipe. Summaries push down to DuckDB.
# work
Everything I have built, in one place: the interactive apps, the packages that sit underneath them, and the databases I have curated.
Filter and subset Tahoe-100M perturbation metadata, 100.6 million rows, and leave with a reproducible pull recipe. Summaries push down to DuckDB.
Shiny app to browse, analyze, and export recount3 RNA-seq studies: 18,998 datasets, quality checks, and PCA.
One gene, one variant, one page. Eighteen cards fan out asynchronously across public APIs, with a 3D structure viewer and a ClinVar typeahead.
Twenty-six pages of GPU-accelerated visualization through shinyreact, in 476 lines of R, one of which is the UI. Precomputed .f32 blobs go straight to the GPU.
Seventeen GPU-accelerated visualizations from one TypeScript core, wrapped three ways: npm for React, htmlwidgets for R, anywidget for Python. The same core drops into all three unchanged.
A gate for biological inputs. Validate gene symbols, ontology terms, variant formats and accessions through one API, offline or live. R and Python share a conformance corpus, so the two cannot disagree.
One client per biological database, each split into a request half and a pure parser half that needs no network to test. Transport and caching are left to biohttp.
Normalized HTTP transport. Calls return a result value rather than raising, so transport failure, a bad status code and an unreadable body stay distinct outcomes. Circuit breaking, retry and caching are defaults.
Reusable, public starter kit for quickly building reproducible bioinformatics and data apps in R Shiny.
Interactive R lessons with paired slide decks, built on the Gapminder dataset for workshops and self-paced learning. Adapted from Software Carpentry's R novice course.
Workshop on deploying a Dockerized R Shiny bioinformatics app on an AWS EC2 instance, using sMAP as the example. Hosted with STEM-Away.
Streamlines project initialization and workflow management.
Manually curated database of 1,700+ glucokinase modulators with bioactivity and chemical data, in-browser 3D structure visualization, and API endpoints.
Manually curated structural database of synthetic peptides containing the non-standard amino acids α,β-dehydrophenylalanine (ΔF) and α-aminoisobutyric acid (Aib).
Resolve a disease name to an EFO/MONDO term, query seven gene-disease sources in parallel, and dedupe to one row per gene. Re-ranks live as you tune the source weights.
Local Shiny app for reviewing Markdown drafts in the browser: paragraph-anchored comments that survive edits, reviews saved as plain YAML, and an optional bring-your-own-key AI assistant. No server, no database, everything is a file.
An agentic evidence-review workbench. Give it a candidate list and a disease context, get back a plausibility-ranked, cited shortlist. Research use only.
Shiny application and R package I built for the Human Genetics team at Genentech for locus-first, integrative visualization across 11,000+ GWAS, eQTL, pQTL, single-cell, and ATAC-seq datasets in Alzheimer's disease.
Record-based Abstraction of Phenotypes, Terms, Ontologies, and Disease Relations. An agentic AI system that pulls phenotypes, genes, diseases, and ontology-linked concepts out of unstructured patient records.
Interactive genomics and clinical-analysis app for whole-exome and bulk RNA-seq pediatric thyroid cancer data.
Online tool to characterize sample subsets (cohorts) and find enriched clinotypes, handy for balancing case/control cohorts and profiling samples in cross-sectional studies.
Shiny app that brings nf-core outputs together with Seurat, pseudobulk, and CellChat for end-to-end single-cell and single-nuclei RNA-seq exploration in one place.
Interactive explorer that combines nf-core whole-exome outputs with variant tools and AI-assisted analysis for NF1-associated tumor models.
R Shiny educational app that walks users through an interactive transcriptomics pipeline with quality control, statistics, and biomarker discovery.
Vitiligo research database with differentially expressed genes, curated protein targets, natural compounds, and co-expression network visualizations.
Pathways, Annotated-lists and Gene-signatures Electronic Repository, with an R Shiny web app for pathway and gene-set enrichment and network interpretation.
R Shiny app that maps and visualizes species occurrences across Poland, using Leaflet for geospatial mapping and dygraphs for time-series exploration.
Chrome extension that restores HTML report viewing in OnDemand v2+. It detects report output and opens it in a clean preview tab instead of the broken in-app rendering.
Every row is something I shipped, every column a skill it uses. Hover a cell to trace it, click a row to open the work.
hover a cell to trace it, click a row to open it