HUB
Sign inSign up

pavlopouloslab/flame

By pavlopouloslab

•Updated 1 day ago

Advanced integration and interpretation of functional enrichment results from multiple sources

Image
0

313

pavlopouloslab/flame repository overview

Docker Pulls License: GPL-3

Flame logo

⁠Flame 3.0: Functional Analysis and Integration made easy

Flame runs functional and literature enrichment on one or many gene lists with eight established tools (g:Profiler, STRING, Enrichr, WebGestalt, PANTHER, GeneCodis, Reactome, clusterProfiler) plus your own GMT gene sets, combines their results, and lets you explore them with interactive plots and networks. It also starts from variants and GWAS results. It is free and open to everyone.

This image bundles the web app and the versioned REST API behind a single port — no separate database or reverse proxy needed to get started.

⁠Features

  • Inputs: paste/upload IDs (19 formats) or ranked lists with scores; select genes on a volcano plot (draggable thresholds), a 2D plot (UMAP/t-SNE/PCA) or an expression matrix (co-expression clusters, gene map, two-group test, pathway activity per sample); map SNPs to genes; find genes in text; microRNAs → miRDB targets; UpSet set operations; and fetch gene lists from a disease (Open Targets, DisGeNET, OMIM, Orphanet, ClinVar, GWAS Catalog), PubMed articles, GEO datasets (RummaGEO) or dbSNP.
  • Enrichment: eight tools across 14,000+ organisms; over-representation (ORA) and, with clusterProfiler/fgsea on MSigDB, gene set enrichment (GSEA) of ranked lists; literature enrichment (PubMed, GeneSigDB, Rummagene); cell types; custom GMT sources, custom backgrounds, per-tool settings.
  • Beyond term lists: transcription factors (ChEA3) and kinases (KEA3), drug and CRISPR signatures that mimic or reverse a response (LINCS L1000), metabolite and lipid pathways (KEGG), STRING network modules with module enrichment, and robustness checks (bootstrap stability, random-list control, FDR thresholds).
  • Results: table, dot plot, bar chart, scatter, heatmaps, networks, Sankey, chord, hive and Circos (edge bundling) views; functional themes and word clouds; a selection panel with ranked, clickable genes, GeneCards and genome browser links, and KEGG maps with the user's genes coloured; undo/redo; image export of every figure (PNG, SVG, PDF, JPEG, WebP) and CSV/HTML exports.
  • Comparison: compare tools, lists or strategies with Fisher's combined p-value, a consensus rank (Robust Rank Aggregation) and a rank chart, a table of which run found what, agreement statistics (Jaccard, rank correlation, McNemar / Cochran's Q, concordance at the top, MDS), UpSet, chord, Circos, bubble plot and heatmap.
  • Variants and GWAS: rsIDs, VCF lines, HGVS or GWAS summary statistics annotated with Ensembl VEP (consequence, impact, gnomAD frequency, ClinVar) and trait associations; Manhattan and genome Circos plots; genes → enrichment.
  • Networks: STRING networks as multi-edge evidence networks; open networks, comparisons and variant reports in Arena3D (3D multilayer networks) or NORMA, or directly in STRING.
  • Display: light/dark themes; every visualization has its own palette (Flame, colour-blind safe, vivid, pastel, grayscale); box and lasso selection on every chart; works at phone width.
  • Workspaces: private bookmarkable links, session save/load (lists, runs, analyses), automatic expiry.
  • Programmatic access: versioned REST API (/api/v1) with an OpenAPI 3 document and Swagger UI, and a Python command-line client.

⁠Quick start

docker pull pavlopouloslab/flame:3.0
docker run -d --name flame -p 8000:8000 -v flame-data:/data pavlopouloslab/flame:3.0

Or with Compose:

services:
  flame:
    image: pavlopouloslab/flame:3.0
    ports:
      - "8000:8000"
    volumes:
      - flame-data:/data
    restart: unless-stopped
volumes:
  flame-data:

Then open http://localhost:8000⁠. The same server provides the web app and the versioned REST API (/api/v1/...; interactive reference at /__docs__/, OpenAPI 3 specification at /openapi.json). Workspaces are stored in the flame-data volume (/data/flame.sqlite), so they survive container upgrades. Pin a version tag (e.g. 3.0, rather than latest) for reproducible deployments.

This image is built for the main path (FLAME_BASE_PATH=/). For a subpath deployment, another R release, or any other build-from-source option, see the GitHub repository⁠.

⁠Configuration (environment variables)

Set any of these with -e VAR=value on docker run, or under environment: in Compose.

VariableDefaultPurpose
FLAME_PORT, FLAME_HOST8000, 0.0.0.0Listening address inside the container
FLAME_DB/data/flame.sqliteWorkspace database
FLAME_WORKERS4Background enrichment workers
FLAME_RETENTION_DAYS30Delete workspaces after this many days without use
FLAME_MAX_LISTS10Lists per workspace (0 = unlimited)
FLAME_PUBLIC_URL–Public base URL (used in links and emails)
FLAME_SMTP_URL, FLAME_SMTP_FROM–Optional "results ready" emails
FLAME_WORKSPACE_LIMIT30New workspaces per hour per client address (0 = no limit)
FLAME_RATE_LIMIT600API requests per minute per client address (0 = no limit, e.g. for local testing)
FLAME_NCBI_API_KEY–NCBI E-utilities key for the PubMed and dbSNP inputs (10 instead of 3 requests per second; recommended for public servers)

⁠Deployment

Run the container behind an HTTPS reverse proxy (nginx, Apache, Caddy, …) on port 443 and forward to FLAME_PORT (8000 by default). The server sets security headers and a strict Content-Security-Policy; the web app loads no third-party scripts, fonts or cookies.

⁠Learn more

Source code, the full configuration reference, building from source, and tests: the GitHub repository⁠.

⁠Publications

If you use Flame, please cite:

  • Karatzas E., Baltoumas F.A., Aplakidou E., Kontou P.I., Stathopoulos P., Stefanis L., Bagos P.G., Pavlopoulos G.A. (2023) Flame (v2.0): advanced integration and interpretation of functional enrichment results from multiple sources. Bioinformatics 39(8):btad490. doi: 10.1093/bioinformatics/btad490⁠
  • Thanati F., Karatzas E., Baltoumas F.A., Stravopodis D.J., Eliopoulos A.G., Pavlopoulos G.A. (2021) FLAME: a web tool for functional and literature enrichment analysis of multiple gene lists. Biology 10(7):665. doi: 10.3390/biology10070665⁠

⁠License

This project is licensed under the GNU General Public License v3.0 (GPL-3) – see the LICENSE⁠ file for details.

Tag summary

Content type

Image

Digest

sha256:cdc35909b…

Size

711.1 MB

Last updated

1 day ago

docker pull pavlopouloslab/flame