GraphSol examples
A single FASTA submission, the API request body, and how to read the id, probability, call and warnings columns in results.csv against summary.json.
These copy-and-adapt patterns walk a GraphSol job end to end — a single FASTA submission, an API request body, and how to read the probability, call, and warnings columns in the results.
GraphSol examples
Use these patterns to prepare inputs, submit jobs, and read outputs quickly.
Single FASTA submission
- Keep headers short and unique.
- Upload as
.fastaor.fasta.gz. - Default threshold is
0.5; adjust if you need precision or recall tuned.
API submission body
- Poll
GET /api/v1/jobs/{jobId}untilsucceeded. - Download
results.csvfor probabilities and calls.
Reading results
id: from FASTA header.probability: 0–1 solubility score.call:soluble/insolubleusing current threshold.warnings: shows trimmed residues or deduplication notes.
Quick quality check
- Compare
summary.jsoninputCountvs your FASTA count. - Scan
warningscolumn for trimmed or invalid residues. - If many sequences cluster near 0.5, consider adjusting threshold offline instead of resubmitting.
Was this page helpful?
Running GraphSol via API or CLI
Submit a graphsol job with a FASTA payload, a threshold, batchSize or skipDeduplication, poll every 5-10 seconds, then pull results.csv or results.json.
Preparing GraphSol inputs
Supply FASTA, UniProt accessions or a previous job output — 20 to 4,096 residues each, 25 MB per upload, the 20 standard amino acids, unique identifiers.