Clann BLAST Explorer

About Clann BLAST Explorer

Clann BLAST Explorer is a free, open-source tool for exploring BLAST and DIAMOND tabular output entirely in your web browser. Open a -outfmt 6 or -outfmt 7 file — or paste tabular text — and see, for each query, a ranked hit table, a best-hit summary, and a coverage diagram of where each high-scoring pair (HSP) falls along the query. Across the whole run, filter by identity, e-value, bit score, alignment length, and query coverage — including by taxonomic rank once a taxonomy source is loaded; view distribution charts and a taxonomy breakdown when taxonomy columns are present; compute reciprocal best hits from a forward/reverse pair; and export FASTA subsets or CSV/TSV tables for downstream work, including alignment and tree building in Clann Tree Viewer and metabarcoding/eDNA analysis in Clann eDNA Explorer.

A one-click built-in taxonomy database — a prebuilt NCBI taxonomy lookup, downloaded once and cached in your browser — resolves full lineage (superkingdom/domain through species) for any hit with a staxids value, no manual taxdump upload required.

Because all parsing and rendering happen locally in your browser, your BLAST results are never uploaded to a server.

Using the explorer

Loading data

Click “Open file…”, drag a file onto the window, or paste tabular text directly. Accepts -outfmt 6 (headerless or with a plain header row), -outfmt 7 (with a "# Fields:" line), DIAMOND's default tabular output, and .gz/.zip compressed versions of any of these — decompressed in your browser before parsing. If a file has no header at all, a column-mapping step appears with a data preview and best-guess dropdowns so you can confirm or correct which column is which before loading, rather than the order being assumed silently.

Per-query view

Pick a query from the “Query” dropdown to see its best hit (or every hit tied for the top bit score, shown separately rather than picking one arbitrarily), a coverage diagram showing where each high-scoring pair (HSP) falls along the query — click a segment to jump straight to its row in the hit table below — and the full ranked, sortable hit table for that query.

All-queries view

Switch to “All queries” at the top of the main panel for a whole-run perspective: a run summary, a per-query summary table (click a row to jump into that query's detail view), a flat “All hits” table spanning every query at once (sortable, toggle between all hits and best-hit-only), distribution charts for %identity/e-value/bit score/alignment length/query coverage, an identity-vs-coverage scatter (colourable by e-value or taxon), a taxonomy breakdown, and RBH mode's results once a reverse run is loaded.

Filters

Set thresholds for %identity, e-value, bit score, alignment length, and query coverage; exclude self-hits; cap hits per query to the top N; and include/exclude specific taxa once taxonomy information is available. “Match rank” scopes that include/exclude to a specific taxonomic rank (superkingdom/domain through species) instead of matching any name or ID text — this only takes effect on hits enriched with full lineage via the built-in taxonomy database; a manual names.dmp upload only carries a flat name, so those hits never match a rank-specific filter. Filters apply live everywhere at once — the per-query view, run summary, charts, and RBH classification all update immediately. Undo steps back through recent changes one at a time; Reset clears everything back to no filtering.

Taxonomy mapping

If your BLAST run has staxids but no scientific names (common without NCBI's taxdb installed locally), two ways to fill them in. Click “Load built-in taxonomy database” for a prebuilt NCBI taxonomy lookup — fetched once (a few tens of MB) and cached in your browser, so every later visit loads instantly with no re-download — which resolves the full lineage (superkingdom/domain through species) for every hit with a staxids value, feeding both the taxonomy chart and the rank-specific filters above. Or upload names.dmp or the full taxdump.tar.gz/new_taxdump.tar.gz yourself — extracted and parsed entirely in your browser — which only backfills a flat scientific/common name (no rank hierarchy). Either way, the taxon ID can come from the staxids column, or be extracted from sseqid itself for ID schemes like STRING/EggNOG's 9606.ENSP00000269305 (a genuine taxon ID as a prefix); a preview always shows exactly what will resolve before you apply the manual-upload path, since a pattern-extracted ID can coincidentally match an unrelated real taxon without looking wrong. “Clear mapping” reverts to the originally parsed data.

RBH mode

Load a second (reverse) BLAST/DIAMOND run to compute reciprocal best hits against the current (forward) run. Each query is classified as reciprocal, one-way, or no hit under the active filters, with a count breakdown, a forward-vs-reverse %identity scatter for reciprocal pairs, and an exportable pair table — all shown in All-queries view.

Export

Choose a scope — the current query, or all hits passing the active filters — then: download the hit table as TSV/CSV; export FASTA directly from qseq/sseq columns when present; upload the original query and/or subject FASTA for ID-matched export of a specific subset (queries passing filters, weak-hit, no-hit, or all queries); combine a query's sequence with its hit sequences into one FASTA ready for an alignment tool; download a plain accession list for NCBI Batch Entrez when sequences aren't available locally; or download an eDNA Explorer sample — a "Lineage TSV" (one row per resolved taxon path, with a name and taxid at every rank when the built-in taxonomy database was used) that treats this run as one metabarcoding sample, ready to load straight into Clann eDNA Explorer for rank views, sunburst/Sankey diagrams, and multi-sample comparison alongside other samples processed the same way.

Frequently asked questions

How do I view BLAST tabular output online?

Open Clann BLAST Explorer, then click “Open file…” or drag your -outfmt 6 or -outfmt 7 tabular file onto the window. You can also paste tabular text straight in. The hit table renders instantly and stays on your machine.

Does it support DIAMOND output?

Yes. DIAMOND's default tabular output uses the same 12-column layout as BLAST -outfmt 6 and is recognised automatically.

Can I upload a gzipped or zipped BLAST file?

Yes — .gz and .zip files are decompressed in your browser before parsing, so you don't need to extract them yourself first. Nothing is uploaded, even temporarily.

What if my file has no header row?

Headerless tabular data is only ever auto-assigned the standard 12-column order when a "# Fields:" line or a recognisable header row says so explicitly — never guessed silently, since a custom -outfmt "6 ..." column order could easily still have exactly 12 columns arranged differently. Otherwise a manual column-mapping step opens: a preview of your first few rows with a dropdown per column, prefilled with a best guess you can correct.

What's the difference between Per-query and All-queries view?

Per-query view shows one query at a time in detail — its best hit, HSP coverage diagram, and full hit table. All-queries view steps back for a whole-run perspective: summary tables, distribution charts, taxonomy, and RBH results across every query at once. Switch between them with the toggle at the top of the main panel; clicking a row in either summary table jumps you into that query's Per-query view.

Can I filter by taxon, identity, or e-value?

Yes, in the Filters panel: thresholds for %identity, e-value, bit score, alignment length, and query coverage; a self-hit exclusion toggle; a top-N-per-query cap; and taxon include/exclude once taxonomy information is available (from the file itself, an uploaded taxonomy mapping, or the built-in taxonomy database). “Match rank” scopes include/exclude to a specific taxonomic rank (superkingdom/domain through species) rather than any name/ID text — this needs the full lineage the built-in taxonomy database resolves, not just a flat name. Every filter applies live across the whole app, with Undo and Reset.

My BLAST output has taxon IDs but no species names — can this fix that?

Yes, two ways. Click “Load built-in taxonomy database” in the Taxonomy mapping panel for a prebuilt NCBI lookup, fetched once and cached in your browser, which resolves full lineage (not just a name) for any staxids value — this also unlocks rank-specific filtering. Or upload NCBI's names.dmp (or the full taxdump.tar.gz) yourself, which backfills scientific/common names wherever a staxids column — or a taxon ID embedded in sseqid itself, for ID schemes like STRING/EggNOG — is present but unnamed (a preview shows exactly what will resolve before you apply it, but no rank hierarchy comes with this path).

What's the built-in taxonomy database, and is it up to date?

A compact binary lookup built from NCBI's taxdump (tools/build-taxonomy-db.js in the repository), served from the same origin as the app itself and fetched once per browser — a few tens of MB, cached in IndexedDB afterwards, so every later visit loads instantly. “Check for updates…” compares your cached copy's build date against the currently published one without re-downloading unless you confirm. It's rebuilt periodically, not live-synced to NCBI.

Can I use this with eDNA/metabarcoding data?

Yes — if you're BLASTing or DIAMOND-ing long reads (e.g. Nanopore amplicons) against a database like NR, load the built-in taxonomy database, filter to your preferred cutoffs, then use “Download eDNA Explorer sample” in the Export panel. It treats the run as one metabarcoding sample (each query = one read, assigned to its best passing hit's taxon) and produces a "Lineage TSV" — full rank hierarchy, not just a name — that loads directly into Clann eDNA Explorer for sunburst/Sankey views and multi-sample comparison. Process each of several samples this way, then load them all into eDNA Explorer together. See a worked example in the repository.

Is my data uploaded to a server?

No. Everything runs client-side in your browser, so your BLAST results never leave your computer.

Can it compute reciprocal best hits (RBH)?

Yes. Load a forward and reverse BLAST run in the RBH mode panel and it classifies each query as reciprocal, one-way, or no hit under the active thresholds, with a pair table and forward-vs-reverse identity scatter.

Can I export sequences for alignment?

Yes, when qseq/sseq columns are present, or by uploading the original query and/or subject FASTA for ID-matched export of the current filtered selection — including a combined query+hits FASTA ready for an alignment tool, and an accession list for NCBI Batch Entrez when sequences aren't available locally.

Does it run BLAST for me?

No. It does not run BLAST or DIAMOND searches, does not fetch sequences from NCBI/UniProt/EBI automatically, and does not build alignments or trees itself — it stages data for tools that do, including Clann Tree Viewer for alignment/phylogenetics and Clann eDNA Explorer for metabarcoding/community analysis.

Is it free?

Yes — Clann BLAST Explorer is free and open source, developed by CreeveyLab. Use it online or clone the repository and host it yourself.

Developed by CreeveyLab · source on GitHub · companion to Clann Tree Viewer and Clann eDNA Explorer.

Part of HoloR-Tools from the HoloRuminant project. This tool was developed with financial support from the European Union's Horizon 2020 research and innovation programme under grant agreement N° 101000213-HoloRuminant. This publication reflects the views only of the author, and not the European Commission (EC); the EC is not liable for any use that may be made of the information contained herein.