Skip to content

Limits and retention

These limits reflect the current code. MB-labelled CRISPRPrep and SpacerTarget input limits use 1,048,576 bytes per MB. A hosting proxy can impose stricter upload limits. An accepted job can wait in a queue before its analysis begins; a busy response without a Job ID requires a later resubmission.

Tool / area Limit
SeqFetcher 500 unique IDs; 16,000 submitted ID-text characters; ID-list file up to 1 MB
SeqFetcher requests At most 5 IDs per request; assembly-containing lists automatically use 1
SeqFetcher response 8 million residues per batch, including all contigs of an assembly
SeqFetcher saved FASTA 25 million characters per job, including headers/newlines
NCBI assembly package 64 MiB downloaded or expanded package size
CRISPRPrep input 100 MB combined; up to 100 uploaded FASTA or imported JSON files; pasted FASTA up to 100 MB
CRISPRPrep NCBI accession input 100 unique nucleotide/assembly IDs and 16,000 text characters; combined retrieved FASTA up to 100 MB
CRISPRPrep extraction 100,000 arrays or extracted spacer rows
CRISPRPrep minimum-spacer option Integer 1–10,000; default 2
CRISPR detection runtime Up to 2 hours for detection; queue waiting can add elapsed time. The service may stop a run earlier
CRISPRPhylo runtime 15 minutes for inference; queue waiting is separate
CRISPRPhylo repeat-distance option Nonnegative integer; default 0
SpacerTarget uploaded input 100 MB combined, up to 100 files
SpacerTarget pasted input 2 MB
SpacerTarget sequence format Each spacer 15–100 A/C/G/T bases
SpacerTarget queries / occurrences 100 distinct sequences / 10,000 occurrences
SpacerTarget search batches At most 10 distinct sequences and 1,000 bases per batch
SpacerTarget target limit 20, 50 (default), or 100 targets per spacer
SpacerTarget network Up to 200 displayed nodes; tables/exports retain all filtered rows
SpacerTarget submitted search A batch exceeding 24 hours is marked failed
ArrayPhyloViz Newick One tree, up to 1 MB
ArrayPhyloViz tree 1,000 leaves, 4,000 nodes, depth 250
ArrayPhyloViz array text 4 MB
ArrayPhyloViz arrays 1,500 arrays; 2,000 spacers per array; 20,000 total instances
ArrayPhyloViz / CRISPRPhylo PNG 32 million pixels and at most 16,000 pixels per side
SpacerMap raster DPI 36–2,400; default 300
SpacerMap occurrence chart 1–50 distinct IDs; above 50, chart preview and chart export are disabled, without truncation
SpacerMap occurrence table / CSV First 100 IDs on page; all IDs in statistics CSV
SpacerMap occurrence raster export 40 million pixels; lower DPI or choose SVG/PDF
My Jobs bulk link import 100 links/IDs per import
Saved results and task records Seven days after the task ends

Queue capacity

Jobs may wait when the service is busy. If a Job ID has been issued, keep the results link and wait for the existing job. If submission is rejected before a Job ID appears, retry after the interval shown on screen. Queue time depends on workload and dataset size.

An accepted Waiting for resources job keeps its input while queued and has no result-expiry timestamp until it ends. Keep its ID instead of repeatedly resubmitting. A 503 busy response before an ID is issued means the submission was rejected; retry after the displayed interval . See My Jobs.

What survives closing the page?

Once a Job ID appears, the accepted analysis continues on the server. Save its private link and retrieve it through My Jobs. The default browser history lasts only for the tab; Remember job history on this device preserves references locally but does not extend server retention.

Original input workspaces are deleted when a task ends. Saved outputs are not a complete archive of uploads, settings or logs. Keep those separately. Further ArrayPhyloViz edits need Save changes or a downloaded project; CRISPRPhylo rerooting changes only the current view/download, not the saved original tree.

For larger data, split analyses into biologically meaningful groups and use SVG for large figures. Download important results before expiry; opening a result link does not reset its expiry time.